Note: Environment variable`HF_TOKEN` is set and is the current active token independently from the token you've just configured. logs/2026-08-01_023219-d24.txt Model parameters: 1,384,122,122 | FLOPs/token: 4.860160e+09 GPU: NVIDIA H100 80GB HBM3 | Peak FLOPS (BF16): 9.89e+14 Total batch size: 1,048,576 tokens = 65,536 tokens/micro x 8 ranks x 2 grad accum step:0/5568 val_bpb:3.175835 val_time:7.11s step 00000/05568 (0.00%) | loss: 10.397411 | lrm: 0.03 | dt: 7183.48ms | tok/sec: 145,970 | bf16_mfu: 8.97 | total time: 0.00m step 00001/05568 (0.02%) | loss: 10.371817 | lrm: 0.05 | dt: 1192.19ms | tok/sec: 879,534 | bf16_mfu: 54.03 | total time: 0.00m step 00002/05568 (0.04%) | loss: 10.325925 | lrm: 0.07 | dt: 1200.06ms | tok/sec: 873,766 | bf16_mfu: 53.67 | total time: 0.00m step 00003/05568 (0.05%) | loss: 10.253621 | lrm: 0.10 | dt: 1200.92ms | tok/sec: 873,140 | bf16_mfu: 53.64 | total time: 0.00m step 00004/05568 (0.07%) | loss: 10.154441 | lrm: 0.12 | dt: 1205.90ms | tok/sec: 869,535 | bf16_mfu: 53.41 | total time: 0.00m step 00005/05568 (0.09%) | loss: 10.002039 | lrm: 0.15 | dt: 1201.81ms | tok/sec: 872,496 | bf16_mfu: 53.60 | total time: 0.00m step 00006/05568 (0.11%) | loss: 9.758124 | lrm: 0.17 | dt: 1200.40ms | tok/sec: 873,524 | bf16_mfu: 53.66 | total time: 0.00m step 00007/05568 (0.13%) | loss: 9.536768 | lrm: 0.20 | dt: 1207.85ms | tok/sec: 868,135 | bf16_mfu: 53.33 | total time: 0.00m step 00008/05568 (0.14%) | loss: 9.184251 | lrm: 0.23 | dt: 1209.30ms | tok/sec: 867,096 | bf16_mfu: 53.26 | total time: 0.00m step 00009/05568 (0.16%) | loss: 8.851041 | lrm: 0.25 | dt: 1203.09ms | tok/sec: 871,566 | bf16_mfu: 53.54 | total time: 0.00m step 00010/05568 (0.18%) | loss: 8.571854 | lrm: 0.28 | dt: 1202.49ms | tok/sec: 872,001 | bf16_mfu: 53.57 | total time: 0.00m step 00011/05568 (0.20%) | loss: 8.322517 | lrm: 0.30 | dt: 1211.17ms | tok/sec: 865,753 | bf16_mfu: 53.18 | total time: 0.02m | eta: 112.2m step 00012/05568 (0.22%) | loss: 8.088904 | lrm: 0.33 | dt: 1204.57ms | tok/sec: 870,495 | bf16_mfu: 53.47 | total time: 0.04m | eta: 111.8m step 00013/05568 (0.23%) | loss: 7.889219 | lrm: 0.35 | dt: 1208.79ms | tok/sec: 867,456 | bf16_mfu: 53.29 | total time: 0.06m | eta: 111.8m step 00014/05568 (0.25%) | loss: 7.687277 | lrm: 0.38 | dt: 1198.23ms | tok/sec: 875,104 | bf16_mfu: 53.76 | total time: 0.08m | eta: 111.6m step 00015/05568 (0.27%) | loss: 7.510805 | lrm: 0.40 | dt: 1204.98ms | tok/sec: 870,199 | bf16_mfu: 53.45 | total time: 0.10m | eta: 111.6m step 00016/05568 (0.29%) | loss: 7.354747 | lrm: 0.42 | dt: 1208.92ms | tok/sec: 867,362 | bf16_mfu: 53.28 | total time: 0.12m | eta: 111.6m step 00017/05568 (0.31%) | loss: 7.202402 | lrm: 0.45 | dt: 1206.11ms | tok/sec: 869,384 | bf16_mfu: 53.40 | total time: 0.14m | eta: 111.6m step 00018/05568 (0.32%) | loss: 7.065530 | lrm: 0.47 | dt: 1205.18ms | tok/sec: 870,060 | bf16_mfu: 53.45 | total time: 0.16m | eta: 111.5m step 00019/05568 (0.34%) | loss: 6.946700 | lrm: 0.50 | dt: 1207.46ms | tok/sec: 868,417 | bf16_mfu: 53.34 | total time: 0.18m | eta: 111.5m step 00020/05568 (0.36%) | loss: 6.826582 | lrm: 0.53 | dt: 1210.43ms | tok/sec: 866,285 | bf16_mfu: 53.21 | total time: 0.20m | eta: 111.5m step 00021/05568 (0.38%) | loss: 6.736403 | lrm: 0.55 | dt: 1208.90ms | tok/sec: 867,377 | bf16_mfu: 53.28 | total time: 0.22m | eta: 111.5m step 00022/05568 (0.40%) | loss: 6.641708 | lrm: 0.57 | dt: 1210.01ms | tok/sec: 866,581 | bf16_mfu: 53.23 | total time: 0.24m | eta: 111.6m step 00023/05568 (0.41%) | loss: 6.549813 | lrm: 0.60 | dt: 1208.03ms | tok/sec: 868,007 | bf16_mfu: 53.32 | total time: 0.26m | eta: 111.5m step 00024/05568 (0.43%) | loss: 6.470866 | lrm: 0.62 | dt: 1211.00ms | tok/sec: 865,876 | bf16_mfu: 53.19 | total time: 0.28m | eta: 111.5m step 00025/05568 (0.45%) | loss: 6.407842 | lrm: 0.65 | dt: 1207.34ms | tok/sec: 868,500 | bf16_mfu: 53.35 | total time: 0.30m | eta: 111.5m step 00026/05568 (0.47%) | loss: 6.334124 | lrm: 0.68 | dt: 1204.73ms | tok/sec: 870,385 | bf16_mfu: 53.47 | total time: 0.32m | eta: 111.5m step 00027/05568 (0.48%) | loss: 6.279915 | lrm: 0.70 | dt: 1209.74ms | tok/sec: 866,776 | bf16_mfu: 53.24 | total time: 0.34m | eta: 111.5m step 00028/05568 (0.50%) | loss: 6.222139 | lrm: 0.72 | dt: 1208.12ms | tok/sec: 867,943 | bf16_mfu: 53.32 | total time: 0.36m | eta: 111.5m step 00029/05568 (0.52%) | loss: 6.171798 | lrm: 0.75 | dt: 1202.33ms | tok/sec: 872,120 | bf16_mfu: 53.57 | total time: 0.38m | eta: 111.4m step 00030/05568 (0.54%) | loss: 6.123902 | lrm: 0.78 | dt: 1203.10ms | tok/sec: 871,562 | bf16_mfu: 53.54 | total time: 0.40m | eta: 111.4m step 00031/05568 (0.56%) | loss: 6.083382 | lrm: 0.80 | dt: 1204.32ms | tok/sec: 870,681 | bf16_mfu: 53.48 | total time: 0.42m | eta: 111.4m step 00032/05568 (0.57%) | loss: 6.038741 | lrm: 0.82 | dt: 1210.66ms | tok/sec: 866,116 | bf16_mfu: 53.20 | total time: 0.44m | eta: 111.3m step 00033/05568 (0.59%) | loss: 5.998780 | lrm: 0.85 | dt: 1208.17ms | tok/sec: 867,906 | bf16_mfu: 53.31 | total time: 0.46m | eta: 111.3m step 00034/05568 (0.61%) | loss: 5.959593 | lrm: 0.88 | dt: 1206.16ms | tok/sec: 869,353 | bf16_mfu: 53.40 | total time: 0.48m | eta: 111.3m step 00035/05568 (0.63%) | loss: 5.926370 | lrm: 0.90 | dt: 1204.49ms | tok/sec: 870,557 | bf16_mfu: 53.48 | total time: 0.50m | eta: 111.3m step 00036/05568 (0.65%) | loss: 5.898817 | lrm: 0.93 | dt: 1200.23ms | tok/sec: 873,646 | bf16_mfu: 53.67 | total time: 0.52m | eta: 111.2m step 00037/05568 (0.66%) | loss: 5.859313 | lrm: 0.95 | dt: 1206.91ms | tok/sec: 868,813 | bf16_mfu: 53.37 | total time: 0.54m | eta: 111.2m step 00038/05568 (0.68%) | loss: 5.826590 | lrm: 0.97 | dt: 1205.83ms | tok/sec: 869,592 | bf16_mfu: 53.42 | total time: 0.56m | eta: 111.2m step 00039/05568 (0.70%) | loss: 5.794246 | lrm: 1.00 | dt: 1208.54ms | tok/sec: 867,636 | bf16_mfu: 53.30 | total time: 0.58m | eta: 111.2m step 00040/05568 (0.72%) | loss: 5.764700 | lrm: 1.00 | dt: 1211.08ms | tok/sec: 865,821 | bf16_mfu: 53.19 | total time: 0.60m | eta: 111.2m step 00041/05568 (0.74%) | loss: 5.738501 | lrm: 1.00 | dt: 1206.54ms | tok/sec: 869,076 | bf16_mfu: 53.39 | total time: 0.62m | eta: 111.2m step 00042/05568 (0.75%) | loss: 5.714134 | lrm: 1.00 | dt: 1205.06ms | tok/sec: 870,141 | bf16_mfu: 53.45 | total time: 0.64m | eta: 111.1m step 00043/05568 (0.77%) | loss: 5.688259 | lrm: 1.00 | dt: 1210.08ms | tok/sec: 866,537 | bf16_mfu: 53.23 | total time: 0.66m | eta: 111.1m step 00044/05568 (0.79%) | loss: 5.667091 | lrm: 1.00 | dt: 1210.41ms | tok/sec: 866,295 | bf16_mfu: 53.21 | total time: 0.68m | eta: 111.1m step 00045/05568 (0.81%) | loss: 5.636713 | lrm: 1.00 | dt: 1206.69ms | tok/sec: 868,967 | bf16_mfu: 53.38 | total time: 0.70m | eta: 111.1m step 00046/05568 (0.83%) | loss: 5.617122 | lrm: 1.00 | dt: 1210.14ms | tok/sec: 866,493 | bf16_mfu: 53.23 | total time: 0.72m | eta: 111.1m step 00047/05568 (0.84%) | loss: 5.587148 | lrm: 1.00 | dt: 1209.43ms | tok/sec: 867,000 | bf16_mfu: 53.26 | total time: 0.74m | eta: 111.1m step 00048/05568 (0.86%) | loss: 5.559172 | lrm: 1.00 | dt: 1208.13ms | tok/sec: 867,934 | bf16_mfu: 53.32 | total time: 0.76m | eta: 111.0m step 00049/05568 (0.88%) | loss: 5.529038 | lrm: 1.00 | dt: 1212.20ms | tok/sec: 865,020 | bf16_mfu: 53.14 | total time: 0.78m | eta: 111.0m step 00050/05568 (0.90%) | loss: 5.507382 | lrm: 1.00 | dt: 1208.35ms | tok/sec: 867,777 | bf16_mfu: 53.31 | total time: 0.80m | eta: 111.0m step 00051/05568 (0.92%) | loss: 5.473621 | lrm: 1.00 | dt: 1214.00ms | tok/sec: 863,736 | bf16_mfu: 53.06 | total time: 0.83m | eta: 111.0m step 00052/05568 (0.93%) | loss: 5.441891 | lrm: 1.00 | dt: 1208.49ms | tok/sec: 867,677 | bf16_mfu: 53.30 | total time: 0.85m | eta: 111.0m step 00053/05568 (0.95%) | loss: 5.410001 | lrm: 1.00 | dt: 1215.98ms | tok/sec: 862,330 | bf16_mfu: 52.97 | total time: 0.87m | eta: 111.0m step 00054/05568 (0.97%) | loss: 5.383740 | lrm: 1.00 | dt: 1208.39ms | tok/sec: 867,743 | bf16_mfu: 53.30 | total time: 0.89m | eta: 111.0m step 00055/05568 (0.99%) | loss: 5.356586 | lrm: 1.00 | dt: 1210.74ms | tok/sec: 866,060 | bf16_mfu: 53.20 | total time: 0.91m | eta: 111.0m step 00056/05568 (1.01%) | loss: 5.326127 | lrm: 1.00 | dt: 1208.88ms | tok/sec: 867,391 | bf16_mfu: 53.28 | total time: 0.93m | eta: 110.9m step 00057/05568 (1.02%) | loss: 5.294790 | lrm: 1.00 | dt: 1210.69ms | tok/sec: 866,096 | bf16_mfu: 53.20 | total time: 0.95m | eta: 110.9m step 00058/05568 (1.04%) | loss: 5.269719 | lrm: 1.00 | dt: 1211.50ms | tok/sec: 865,522 | bf16_mfu: 53.17 | total time: 0.97m | eta: 110.9m step 00059/05568 (1.06%) | loss: 5.240433 | lrm: 1.00 | dt: 1212.72ms | tok/sec: 864,648 | bf16_mfu: 53.11 | total time: 0.99m | eta: 110.9m step 00060/05568 (1.08%) | loss: 5.213633 | lrm: 1.00 | dt: 1212.97ms | tok/sec: 864,470 | bf16_mfu: 53.10 | total time: 1.01m | eta: 110.9m step 00061/05568 (1.10%) | loss: 5.194553 | lrm: 1.00 | dt: 1216.81ms | tok/sec: 861,741 | bf16_mfu: 52.93 | total time: 1.03m | eta: 110.9m step 00062/05568 (1.11%) | loss: 5.161059 | lrm: 1.00 | dt: 1217.75ms | tok/sec: 861,073 | bf16_mfu: 52.89 | total time: 1.05m | eta: 110.9m step 00063/05568 (1.13%) | loss: 5.139209 | lrm: 1.00 | dt: 1211.76ms | tok/sec: 865,335 | bf16_mfu: 53.16 | total time: 1.07m | eta: 110.9m step 00064/05568 (1.15%) | loss: 5.119437 | lrm: 1.00 | dt: 1211.31ms | tok/sec: 865,658 | bf16_mfu: 53.18 | total time: 1.09m | eta: 110.8m step 00065/05568 (1.17%) | loss: 5.083644 | lrm: 1.00 | dt: 1213.02ms | tok/sec: 864,432 | bf16_mfu: 53.10 | total time: 1.11m | eta: 110.8m step 00066/05568 (1.19%) | loss: 5.058061 | lrm: 1.00 | dt: 1212.32ms | tok/sec: 864,934 | bf16_mfu: 53.13 | total time: 1.13m | eta: 110.8m step 00067/05568 (1.20%) | loss: 5.035833 | lrm: 1.00 | dt: 1214.32ms | tok/sec: 863,508 | bf16_mfu: 53.04 | total time: 1.15m | eta: 110.8m step 00068/05568 (1.22%) | loss: 5.014245 | lrm: 1.00 | dt: 1216.24ms | tok/sec: 862,142 | bf16_mfu: 52.96 | total time: 1.17m | eta: 110.8m step 00069/05568 (1.24%) | loss: 4.985068 | lrm: 1.00 | dt: 1213.41ms | tok/sec: 864,154 | bf16_mfu: 53.08 | total time: 1.19m | eta: 110.8m step 00070/05568 (1.26%) | loss: 4.971687 | lrm: 1.00 | dt: 1215.20ms | tok/sec: 862,879 | bf16_mfu: 53.00 | total time: 1.21m | eta: 110.8m step 00071/05568 (1.28%) | loss: 4.951897 | lrm: 1.00 | dt: 1215.03ms | tok/sec: 863,002 | bf16_mfu: 53.01 | total time: 1.23m | eta: 110.8m step 00072/05568 (1.29%) | loss: 4.929701 | lrm: 1.00 | dt: 1217.38ms | tok/sec: 861,338 | bf16_mfu: 52.91 | total time: 1.25m | eta: 110.8m step 00073/05568 (1.31%) | loss: 4.911270 | lrm: 1.00 | dt: 1217.17ms | tok/sec: 861,483 | bf16_mfu: 52.92 | total time: 1.27m | eta: 110.8m step 00074/05568 (1.33%) | loss: 4.901338 | lrm: 1.00 | dt: 1217.29ms | tok/sec: 861,401 | bf16_mfu: 52.91 | total time: 1.29m | eta: 110.7m step 00075/05568 (1.35%) | loss: 4.882687 | lrm: 1.00 | dt: 1219.36ms | tok/sec: 859,937 | bf16_mfu: 52.82 | total time: 1.31m | eta: 110.7m step 00076/05568 (1.36%) | loss: 4.866796 | lrm: 1.00 | dt: 1215.36ms | tok/sec: 862,766 | bf16_mfu: 53.00 | total time: 1.33m | eta: 110.7m step 00077/05568 (1.38%) | loss: 4.840702 | lrm: 1.00 | dt: 1215.91ms | tok/sec: 862,382 | bf16_mfu: 52.97 | total time: 1.35m | eta: 110.7m step 00078/05568 (1.40%) | loss: 4.820797 | lrm: 1.00 | dt: 1217.51ms | tok/sec: 861,248 | bf16_mfu: 52.90 | total time: 1.37m | eta: 110.7m step 00079/05568 (1.42%) | loss: 4.802415 | lrm: 1.00 | dt: 1218.04ms | tok/sec: 860,872 | bf16_mfu: 52.88 | total time: 1.39m | eta: 110.7m step 00080/05568 (1.44%) | loss: 4.787924 | lrm: 1.00 | dt: 1215.62ms | tok/sec: 862,584 | bf16_mfu: 52.99 | total time: 1.41m | eta: 110.7m step 00081/05568 (1.45%) | loss: 4.768831 | lrm: 1.00 | dt: 1215.72ms | tok/sec: 862,515 | bf16_mfu: 52.98 | total time: 1.43m | eta: 110.7m step 00082/05568 (1.47%) | loss: 4.756791 | lrm: 1.00 | dt: 1215.95ms | tok/sec: 862,354 | bf16_mfu: 52.97 | total time: 1.45m | eta: 110.7m step 00083/05568 (1.49%) | loss: 4.733536 | lrm: 1.00 | dt: 1216.21ms | tok/sec: 862,163 | bf16_mfu: 52.96 | total time: 1.47m | eta: 110.6m step 00084/05568 (1.51%) | loss: 4.709723 | lrm: 1.00 | dt: 1219.55ms | tok/sec: 859,802 | bf16_mfu: 52.82 | total time: 1.49m | eta: 110.6m step 00085/05568 (1.53%) | loss: 4.692001 | lrm: 1.00 | dt: 1216.83ms | tok/sec: 861,728 | bf16_mfu: 52.93 | total time: 1.51m | eta: 110.6m step 00086/05568 (1.54%) | loss: 4.674702 | lrm: 1.00 | dt: 1214.17ms | tok/sec: 863,617 | bf16_mfu: 53.05 | total time: 1.53m | eta: 110.6m step 00087/05568 (1.56%) | loss: 4.652666 | lrm: 1.00 | dt: 1219.55ms | tok/sec: 859,807 | bf16_mfu: 52.82 | total time: 1.55m | eta: 110.6m step 00088/05568 (1.58%) | loss: 4.633362 | lrm: 1.00 | dt: 1215.58ms | tok/sec: 862,616 | bf16_mfu: 52.99 | total time: 1.57m | eta: 110.6m step 00089/05568 (1.60%) | loss: 4.625807 | lrm: 1.00 | dt: 1217.36ms | tok/sec: 861,355 | bf16_mfu: 52.91 | total time: 1.59m | eta: 110.6m step 00090/05568 (1.62%) | loss: 4.609017 | lrm: 1.00 | dt: 1217.94ms | tok/sec: 860,938 | bf16_mfu: 52.89 | total time: 1.61m | eta: 110.6m step 00091/05568 (1.63%) | loss: 4.595796 | lrm: 1.00 | dt: 1221.14ms | tok/sec: 858,683 | bf16_mfu: 52.75 | total time: 1.64m | eta: 110.5m step 00092/05568 (1.65%) | loss: 4.571985 | lrm: 1.00 | dt: 1218.46ms | tok/sec: 860,572 | bf16_mfu: 52.86 | total time: 1.66m | eta: 110.5m step 00093/05568 (1.67%) | loss: 4.551725 | lrm: 1.00 | dt: 1217.34ms | tok/sec: 861,369 | bf16_mfu: 52.91 | total time: 1.68m | eta: 110.5m step 00094/05568 (1.69%) | loss: 4.541835 | lrm: 1.00 | dt: 1216.67ms | tok/sec: 861,838 | bf16_mfu: 52.94 | total time: 1.70m | eta: 110.5m step 00095/05568 (1.71%) | loss: 4.527871 | lrm: 1.00 | dt: 1219.26ms | tok/sec: 860,006 | bf16_mfu: 52.83 | total time: 1.72m | eta: 110.5m step 00096/05568 (1.72%) | loss: 4.512148 | lrm: 1.00 | dt: 1224.07ms | tok/sec: 856,630 | bf16_mfu: 52.62 | total time: 1.74m | eta: 110.5m step 00097/05568 (1.74%) | loss: 4.501337 | lrm: 1.00 | dt: 1216.89ms | tok/sec: 861,683 | bf16_mfu: 52.93 | total time: 1.76m | eta: 110.5m step 00098/05568 (1.76%) | loss: 4.486491 | lrm: 1.00 | dt: 1213.48ms | tok/sec: 864,103 | bf16_mfu: 53.08 | total time: 1.78m | eta: 110.5m step 00099/05568 (1.78%) | loss: 4.471021 | lrm: 1.00 | dt: 1217.31ms | tok/sec: 861,389 | bf16_mfu: 52.91 | total time: 1.80m | eta: 110.4m step 00100/05568 (1.80%) | loss: 4.449563 | lrm: 1.00 | dt: 1214.16ms | tok/sec: 863,624 | bf16_mfu: 53.05 | total time: 1.82m | eta: 110.4m step 00101/05568 (1.81%) | loss: 4.441779 | lrm: 1.00 | dt: 1218.40ms | tok/sec: 860,613 | bf16_mfu: 52.87 | total time: 1.84m | eta: 110.4m step 00102/05568 (1.83%) | loss: 4.422543 | lrm: 1.00 | dt: 1223.45ms | tok/sec: 857,064 | bf16_mfu: 52.65 | total time: 1.86m | eta: 110.4m step 00103/05568 (1.85%) | loss: 4.408286 | lrm: 1.00 | dt: 1218.61ms | tok/sec: 860,469 | bf16_mfu: 52.86 | total time: 1.88m | eta: 110.4m step 00104/05568 (1.87%) | loss: 4.391420 | lrm: 1.00 | dt: 1219.13ms | tok/sec: 860,102 | bf16_mfu: 52.83 | total time: 1.90m | eta: 110.4m step 00105/05568 (1.89%) | loss: 4.376288 | lrm: 1.00 | dt: 1218.36ms | tok/sec: 860,642 | bf16_mfu: 52.87 | total time: 1.92m | eta: 110.4m step 00106/05568 (1.90%) | loss: 4.358902 | lrm: 1.00 | dt: 1221.79ms | tok/sec: 858,228 | bf16_mfu: 52.72 | total time: 1.94m | eta: 110.3m step 00107/05568 (1.92%) | loss: 4.336580 | lrm: 1.00 | dt: 1221.94ms | tok/sec: 858,122 | bf16_mfu: 52.71 | total time: 1.96m | eta: 110.3m step 00108/05568 (1.94%) | loss: 4.319047 | lrm: 1.00 | dt: 1217.11ms | tok/sec: 861,532 | bf16_mfu: 52.92 | total time: 1.98m | eta: 110.3m step 00109/05568 (1.96%) | loss: 4.299571 | lrm: 1.00 | dt: 1222.65ms | tok/sec: 857,626 | bf16_mfu: 52.68 | total time: 2.00m | eta: 110.3m step 00110/05568 (1.98%) | loss: 4.275272 | lrm: 1.00 | dt: 1220.11ms | tok/sec: 859,410 | bf16_mfu: 52.79 | total time: 2.02m | eta: 110.3m step 00111/05568 (1.99%) | loss: 4.268009 | lrm: 1.00 | dt: 1214.19ms | tok/sec: 863,602 | bf16_mfu: 53.05 | total time: 2.04m | eta: 110.3m step 00112/05568 (2.01%) | loss: 4.246765 | lrm: 1.00 | dt: 1220.61ms | tok/sec: 859,057 | bf16_mfu: 52.77 | total time: 2.06m | eta: 110.3m step 00113/05568 (2.03%) | loss: 4.242301 | lrm: 1.00 | dt: 1220.18ms | tok/sec: 859,358 | bf16_mfu: 52.79 | total time: 2.08m | eta: 110.2m step 00114/05568 (2.05%) | loss: 4.222721 | lrm: 1.00 | dt: 1213.46ms | tok/sec: 864,123 | bf16_mfu: 53.08 | total time: 2.10m | eta: 110.2m step 00115/05568 (2.07%) | loss: 4.197419 | lrm: 1.00 | dt: 1219.59ms | tok/sec: 859,776 | bf16_mfu: 52.81 | total time: 2.12m | eta: 110.2m step 00116/05568 (2.08%) | loss: 4.178493 | lrm: 1.00 | dt: 1219.87ms | tok/sec: 859,578 | bf16_mfu: 52.80 | total time: 2.14m | eta: 110.2m step 00117/05568 (2.10%) | loss: 4.171298 | lrm: 1.00 | dt: 1222.02ms | tok/sec: 858,065 | bf16_mfu: 52.71 | total time: 2.16m | eta: 110.2m step 00118/05568 (2.12%) | loss: 4.158215 | lrm: 1.00 | dt: 1223.69ms | tok/sec: 856,897 | bf16_mfu: 52.64 | total time: 2.18m | eta: 110.2m step 00119/05568 (2.14%) | loss: 4.147682 | lrm: 1.00 | dt: 1218.22ms | tok/sec: 860,746 | bf16_mfu: 52.87 | total time: 2.20m | eta: 110.2m step 00120/05568 (2.16%) | loss: 4.139920 | lrm: 1.00 | dt: 1221.97ms | tok/sec: 858,103 | bf16_mfu: 52.71 | total time: 2.22m | eta: 110.1m step 00121/05568 (2.17%) | loss: 4.135097 | lrm: 1.00 | dt: 1216.75ms | tok/sec: 861,785 | bf16_mfu: 52.94 | total time: 2.24m | eta: 110.1m step 00122/05568 (2.19%) | loss: 4.119586 | lrm: 1.00 | dt: 1224.41ms | tok/sec: 856,393 | bf16_mfu: 52.61 | total time: 2.27m | eta: 110.1m step 00123/05568 (2.21%) | loss: 4.115093 | lrm: 1.00 | dt: 1222.73ms | tok/sec: 857,572 | bf16_mfu: 52.68 | total time: 2.29m | eta: 110.1m step 00124/05568 (2.23%) | loss: 4.110966 | lrm: 1.00 | dt: 1217.32ms | tok/sec: 861,382 | bf16_mfu: 52.91 | total time: 2.31m | eta: 110.1m step 00125/05568 (2.24%) | loss: 4.095991 | lrm: 1.00 | dt: 1217.15ms | tok/sec: 861,503 | bf16_mfu: 52.92 | total time: 2.33m | eta: 110.1m step 00126/05568 (2.26%) | loss: 4.089274 | lrm: 1.00 | dt: 1219.10ms | tok/sec: 860,125 | bf16_mfu: 52.84 | total time: 2.35m | eta: 110.1m step 00127/05568 (2.28%) | loss: 4.072843 | lrm: 1.00 | dt: 1221.74ms | tok/sec: 858,262 | bf16_mfu: 52.72 | total time: 2.37m | eta: 110.0m step 00128/05568 (2.30%) | loss: 4.047993 | lrm: 1.00 | dt: 1220.15ms | tok/sec: 859,384 | bf16_mfu: 52.79 | total time: 2.39m | eta: 110.0m step 00129/05568 (2.32%) | loss: 4.036010 | lrm: 1.00 | dt: 1219.36ms | tok/sec: 859,940 | bf16_mfu: 52.82 | total time: 2.41m | eta: 110.0m step 00130/05568 (2.33%) | loss: 4.011930 | lrm: 1.00 | dt: 1220.27ms | tok/sec: 859,301 | bf16_mfu: 52.78 | total time: 2.43m | eta: 110.0m step 00131/05568 (2.35%) | loss: 3.998977 | lrm: 1.00 | dt: 1220.49ms | tok/sec: 859,143 | bf16_mfu: 52.78 | total time: 2.45m | eta: 110.0m step 00132/05568 (2.37%) | loss: 3.983711 | lrm: 1.00 | dt: 1222.07ms | tok/sec: 858,031 | bf16_mfu: 52.71 | total time: 2.47m | eta: 110.0m step 00133/05568 (2.39%) | loss: 3.969216 | lrm: 1.00 | dt: 1216.15ms | tok/sec: 862,209 | bf16_mfu: 52.96 | total time: 2.49m | eta: 109.9m step 00134/05568 (2.41%) | loss: 3.956401 | lrm: 1.00 | dt: 1217.94ms | tok/sec: 860,939 | bf16_mfu: 52.89 | total time: 2.51m | eta: 109.9m step 00135/05568 (2.42%) | loss: 3.945336 | lrm: 1.00 | dt: 1218.91ms | tok/sec: 860,256 | bf16_mfu: 52.84 | total time: 2.53m | eta: 109.9m step 00136/05568 (2.44%) | loss: 3.916937 | lrm: 1.00 | dt: 1222.49ms | tok/sec: 857,740 | bf16_mfu: 52.69 | total time: 2.55m | eta: 109.9m step 00137/05568 (2.46%) | loss: 3.913222 | lrm: 1.00 | dt: 1223.86ms | tok/sec: 856,775 | bf16_mfu: 52.63 | total time: 2.57m | eta: 109.9m step 00138/05568 (2.48%) | loss: 3.896956 | lrm: 1.00 | dt: 1217.11ms | tok/sec: 861,526 | bf16_mfu: 52.92 | total time: 2.59m | eta: 109.9m step 00139/05568 (2.50%) | loss: 3.874416 | lrm: 1.00 | dt: 1218.19ms | tok/sec: 860,769 | bf16_mfu: 52.88 | total time: 2.61m | eta: 109.8m step 00140/05568 (2.51%) | loss: 3.864788 | lrm: 1.00 | dt: 1219.80ms | tok/sec: 859,632 | bf16_mfu: 52.81 | total time: 2.63m | eta: 109.8m step 00141/05568 (2.53%) | loss: 3.861869 | lrm: 1.00 | dt: 1217.42ms | tok/sec: 861,307 | bf16_mfu: 52.91 | total time: 2.65m | eta: 109.8m step 00142/05568 (2.55%) | loss: 3.852290 | lrm: 1.00 | dt: 1223.23ms | tok/sec: 857,222 | bf16_mfu: 52.66 | total time: 2.67m | eta: 109.8m step 00143/05568 (2.57%) | loss: 3.838713 | lrm: 1.00 | dt: 1223.17ms | tok/sec: 857,260 | bf16_mfu: 52.66 | total time: 2.69m | eta: 109.8m step 00144/05568 (2.59%) | loss: 3.816625 | lrm: 1.00 | dt: 1211.74ms | tok/sec: 865,344 | bf16_mfu: 53.16 | total time: 2.71m | eta: 109.8m step 00145/05568 (2.60%) | loss: 3.812562 | lrm: 1.00 | dt: 1216.69ms | tok/sec: 861,823 | bf16_mfu: 52.94 | total time: 2.73m | eta: 109.7m step 00146/05568 (2.62%) | loss: 3.789266 | lrm: 1.00 | dt: 1219.13ms | tok/sec: 860,100 | bf16_mfu: 52.83 | total time: 2.75m | eta: 109.7m step 00147/05568 (2.64%) | loss: 3.787553 | lrm: 1.00 | dt: 1218.02ms | tok/sec: 860,885 | bf16_mfu: 52.88 | total time: 2.77m | eta: 109.7m step 00148/05568 (2.66%) | loss: 3.774551 | lrm: 1.00 | dt: 1218.81ms | tok/sec: 860,327 | bf16_mfu: 52.85 | total time: 2.79m | eta: 109.7m step 00149/05568 (2.68%) | loss: 3.762742 | lrm: 1.00 | dt: 1216.74ms | tok/sec: 861,791 | bf16_mfu: 52.94 | total time: 2.81m | eta: 109.7m step 00150/05568 (2.69%) | loss: 3.761492 | lrm: 1.00 | dt: 1220.03ms | tok/sec: 859,469 | bf16_mfu: 52.80 | total time: 2.83m | eta: 109.7m step 00151/05568 (2.71%) | loss: 3.751402 | lrm: 1.00 | dt: 1218.95ms | tok/sec: 860,229 | bf16_mfu: 52.84 | total time: 2.85m | eta: 109.6m step 00152/05568 (2.73%) | loss: 3.750820 | lrm: 1.00 | dt: 1217.00ms | tok/sec: 861,608 | bf16_mfu: 52.93 | total time: 2.87m | eta: 109.6m step 00153/05568 (2.75%) | loss: 3.737993 | lrm: 1.00 | dt: 1226.78ms | tok/sec: 854,739 | bf16_mfu: 52.50 | total time: 2.90m | eta: 109.6m step 00154/05568 (2.77%) | loss: 3.740709 | lrm: 1.00 | dt: 1217.08ms | tok/sec: 861,552 | bf16_mfu: 52.92 | total time: 2.92m | eta: 109.6m step 00155/05568 (2.78%) | loss: 3.731952 | lrm: 1.00 | dt: 1217.76ms | tok/sec: 861,066 | bf16_mfu: 52.89 | total time: 2.94m | eta: 109.6m step 00156/05568 (2.80%) | loss: 3.713869 | lrm: 1.00 | dt: 1213.63ms | tok/sec: 863,999 | bf16_mfu: 53.07 | total time: 2.96m | eta: 109.5m step 00157/05568 (2.82%) | loss: 3.710732 | lrm: 1.00 | dt: 1219.15ms | tok/sec: 860,084 | bf16_mfu: 52.83 | total time: 2.98m | eta: 109.5m step 00158/05568 (2.84%) | loss: 3.709160 | lrm: 1.00 | dt: 1219.24ms | tok/sec: 860,027 | bf16_mfu: 52.83 | total time: 3.00m | eta: 109.5m step 00159/05568 (2.86%) | loss: 3.703904 | lrm: 1.00 | dt: 1220.99ms | tok/sec: 858,789 | bf16_mfu: 52.75 | total time: 3.02m | eta: 109.5m step 00160/05568 (2.87%) | loss: 3.707567 | lrm: 1.00 | dt: 1213.63ms | tok/sec: 864,003 | bf16_mfu: 53.07 | total time: 3.04m | eta: 109.5m step 00161/05568 (2.89%) | loss: 3.693467 | lrm: 1.00 | dt: 1219.43ms | tok/sec: 859,891 | bf16_mfu: 52.82 | total time: 3.06m | eta: 109.5m step 00162/05568 (2.91%) | loss: 3.687130 | lrm: 1.00 | dt: 1217.78ms | tok/sec: 861,052 | bf16_mfu: 52.89 | total time: 3.08m | eta: 109.4m step 00163/05568 (2.93%) | loss: 3.677243 | lrm: 1.00 | dt: 1215.71ms | tok/sec: 862,520 | bf16_mfu: 52.98 | total time: 3.10m | eta: 109.4m step 00164/05568 (2.95%) | loss: 3.660563 | lrm: 1.00 | dt: 1220.68ms | tok/sec: 859,010 | bf16_mfu: 52.77 | total time: 3.12m | eta: 109.4m step 00165/05568 (2.96%) | loss: 3.657155 | lrm: 1.00 | dt: 1215.86ms | tok/sec: 862,411 | bf16_mfu: 52.98 | total time: 3.14m | eta: 109.4m step 00166/05568 (2.98%) | loss: 3.658240 | lrm: 1.00 | dt: 1218.65ms | tok/sec: 860,440 | bf16_mfu: 52.85 | total time: 3.16m | eta: 109.4m step 00167/05568 (3.00%) | loss: 3.650298 | lrm: 1.00 | dt: 1216.16ms | tok/sec: 862,200 | bf16_mfu: 52.96 | total time: 3.18m | eta: 109.3m step 00168/05568 (3.02%) | loss: 3.647943 | lrm: 1.00 | dt: 1217.80ms | tok/sec: 861,041 | bf16_mfu: 52.89 | total time: 3.20m | eta: 109.3m step 00169/05568 (3.04%) | loss: 3.632838 | lrm: 1.00 | dt: 1225.75ms | tok/sec: 855,457 | bf16_mfu: 52.55 | total time: 3.22m | eta: 109.3m step 00170/05568 (3.05%) | loss: 3.632776 | lrm: 1.00 | dt: 1214.60ms | tok/sec: 863,306 | bf16_mfu: 53.03 | total time: 3.24m | eta: 109.3m step 00171/05568 (3.07%) | loss: 3.627982 | lrm: 1.00 | dt: 1219.51ms | tok/sec: 859,833 | bf16_mfu: 52.82 | total time: 3.26m | eta: 109.3m step 00172/05568 (3.09%) | loss: 3.620407 | lrm: 1.00 | dt: 1221.86ms | tok/sec: 858,177 | bf16_mfu: 52.72 | total time: 3.28m | eta: 109.3m step 00173/05568 (3.11%) | loss: 3.611766 | lrm: 1.00 | dt: 1220.18ms | tok/sec: 859,363 | bf16_mfu: 52.79 | total time: 3.30m | eta: 109.2m step 00174/05568 (3.12%) | loss: 3.590008 | lrm: 1.00 | dt: 1223.50ms | tok/sec: 857,027 | bf16_mfu: 52.65 | total time: 3.32m | eta: 109.2m step 00175/05568 (3.14%) | loss: 3.581273 | lrm: 1.00 | dt: 1216.61ms | tok/sec: 861,885 | bf16_mfu: 52.94 | total time: 3.34m | eta: 109.2m step 00176/05568 (3.16%) | loss: 3.565322 | lrm: 1.00 | dt: 1214.73ms | tok/sec: 863,218 | bf16_mfu: 53.03 | total time: 3.36m | eta: 109.2m step 00177/05568 (3.18%) | loss: 3.545206 | lrm: 1.00 | dt: 1221.57ms | tok/sec: 858,383 | bf16_mfu: 52.73 | total time: 3.38m | eta: 109.2m step 00178/05568 (3.20%) | loss: 3.556319 | lrm: 1.00 | dt: 1219.43ms | tok/sec: 859,888 | bf16_mfu: 52.82 | total time: 3.40m | eta: 109.2m step 00179/05568 (3.21%) | loss: 3.553619 | lrm: 1.00 | dt: 1225.24ms | tok/sec: 855,810 | bf16_mfu: 52.57 | total time: 3.42m | eta: 109.1m step 00180/05568 (3.23%) | loss: 3.541996 | lrm: 1.00 | dt: 1220.07ms | tok/sec: 859,436 | bf16_mfu: 52.79 | total time: 3.44m | eta: 109.1m step 00181/05568 (3.25%) | loss: 3.544237 | lrm: 1.00 | dt: 1214.49ms | tok/sec: 863,385 | bf16_mfu: 53.04 | total time: 3.46m | eta: 109.1m step 00182/05568 (3.27%) | loss: 3.533691 | lrm: 1.00 | dt: 1217.53ms | tok/sec: 861,231 | bf16_mfu: 52.90 | total time: 3.48m | eta: 109.1m step 00183/05568 (3.29%) | loss: 3.532685 | lrm: 1.00 | dt: 1214.93ms | tok/sec: 863,076 | bf16_mfu: 53.02 | total time: 3.50m | eta: 109.1m step 00184/05568 (3.30%) | loss: 3.535770 | lrm: 1.00 | dt: 1214.69ms | tok/sec: 863,243 | bf16_mfu: 53.03 | total time: 3.52m | eta: 109.0m step 00185/05568 (3.32%) | loss: 3.525240 | lrm: 1.00 | dt: 1219.24ms | tok/sec: 860,027 | bf16_mfu: 52.83 | total time: 3.54m | eta: 109.0m step 00186/05568 (3.34%) | loss: 3.513576 | lrm: 1.00 | dt: 1216.54ms | tok/sec: 861,930 | bf16_mfu: 52.95 | total time: 3.57m | eta: 109.0m step 00187/05568 (3.36%) | loss: 3.502529 | lrm: 1.00 | dt: 1216.75ms | tok/sec: 861,783 | bf16_mfu: 52.94 | total time: 3.59m | eta: 109.0m step 00188/05568 (3.38%) | loss: 3.495289 | lrm: 1.00 | dt: 1215.71ms | tok/sec: 862,523 | bf16_mfu: 52.98 | total time: 3.61m | eta: 109.0m step 00189/05568 (3.39%) | loss: 3.497829 | lrm: 1.00 | dt: 1219.20ms | tok/sec: 860,051 | bf16_mfu: 52.83 | total time: 3.63m | eta: 108.9m step 00190/05568 (3.41%) | loss: 3.494688 | lrm: 1.00 | dt: 1219.11ms | tok/sec: 860,114 | bf16_mfu: 52.83 | total time: 3.65m | eta: 108.9m step 00191/05568 (3.43%) | loss: 3.488985 | lrm: 1.00 | dt: 1217.82ms | tok/sec: 861,029 | bf16_mfu: 52.89 | total time: 3.67m | eta: 108.9m step 00192/05568 (3.45%) | loss: 3.483672 | lrm: 1.00 | dt: 1212.72ms | tok/sec: 864,650 | bf16_mfu: 53.11 | total time: 3.69m | eta: 108.9m step 00193/05568 (3.47%) | loss: 3.483420 | lrm: 1.00 | dt: 1215.57ms | tok/sec: 862,618 | bf16_mfu: 52.99 | total time: 3.71m | eta: 108.9m step 00194/05568 (3.48%) | loss: 3.493178 | lrm: 1.00 | dt: 1212.29ms | tok/sec: 864,951 | bf16_mfu: 53.13 | total time: 3.73m | eta: 108.8m step 00195/05568 (3.50%) | loss: 3.479336 | lrm: 1.00 | dt: 1209.98ms | tok/sec: 866,603 | bf16_mfu: 53.23 | total time: 3.75m | eta: 108.8m step 00196/05568 (3.52%) | loss: 3.483594 | lrm: 1.00 | dt: 1217.96ms | tok/sec: 860,928 | bf16_mfu: 52.88 | total time: 3.77m | eta: 108.8m step 00197/05568 (3.54%) | loss: 3.491738 | lrm: 1.00 | dt: 1222.23ms | tok/sec: 857,918 | bf16_mfu: 52.70 | total time: 3.79m | eta: 108.8m step 00198/05568 (3.56%) | loss: 3.485522 | lrm: 1.00 | dt: 1216.48ms | tok/sec: 861,975 | bf16_mfu: 52.95 | total time: 3.81m | eta: 108.8m step 00199/05568 (3.57%) | loss: 3.465492 | lrm: 1.00 | dt: 1219.61ms | tok/sec: 859,765 | bf16_mfu: 52.81 | total time: 3.83m | eta: 108.7m step 00200/05568 (3.59%) | loss: 3.460220 | lrm: 1.00 | dt: 1216.81ms | tok/sec: 861,738 | bf16_mfu: 52.93 | total time: 3.85m | eta: 108.7m step 00201/05568 (3.61%) | loss: 3.449767 | lrm: 1.00 | dt: 1213.54ms | tok/sec: 864,065 | bf16_mfu: 53.08 | total time: 3.87m | eta: 108.7m step 00202/05568 (3.63%) | loss: 3.448646 | lrm: 1.00 | dt: 1214.58ms | tok/sec: 863,325 | bf16_mfu: 53.03 | total time: 3.89m | eta: 108.7m step 00203/05568 (3.65%) | loss: 3.450988 | lrm: 1.00 | dt: 1212.02ms | tok/sec: 865,150 | bf16_mfu: 53.14 | total time: 3.91m | eta: 108.7m step 00204/05568 (3.66%) | loss: 3.442147 | lrm: 1.00 | dt: 1215.63ms | tok/sec: 862,575 | bf16_mfu: 52.99 | total time: 3.93m | eta: 108.6m step 00205/05568 (3.68%) | loss: 3.457836 | lrm: 1.00 | dt: 1220.10ms | tok/sec: 859,416 | bf16_mfu: 52.79 | total time: 3.95m | eta: 108.6m step 00206/05568 (3.70%) | loss: 3.453585 | lrm: 1.00 | dt: 1214.19ms | tok/sec: 863,603 | bf16_mfu: 53.05 | total time: 3.97m | eta: 108.6m step 00207/05568 (3.72%) | loss: 3.443596 | lrm: 1.00 | dt: 1211.50ms | tok/sec: 865,515 | bf16_mfu: 53.17 | total time: 3.99m | eta: 108.6m step 00208/05568 (3.74%) | loss: 3.446481 | lrm: 1.00 | dt: 1215.82ms | tok/sec: 862,444 | bf16_mfu: 52.98 | total time: 4.01m | eta: 108.6m step 00209/05568 (3.75%) | loss: 3.442501 | lrm: 1.00 | dt: 1214.46ms | tok/sec: 863,410 | bf16_mfu: 53.04 | total time: 4.03m | eta: 108.5m step 00210/05568 (3.77%) | loss: 3.438154 | lrm: 1.00 | dt: 1211.06ms | tok/sec: 865,833 | bf16_mfu: 53.19 | total time: 4.05m | eta: 108.5m step 00211/05568 (3.79%) | loss: 3.419605 | lrm: 1.00 | dt: 1215.51ms | tok/sec: 862,666 | bf16_mfu: 52.99 | total time: 4.07m | eta: 108.5m step 00212/05568 (3.81%) | loss: 3.412897 | lrm: 1.00 | dt: 1226.16ms | tok/sec: 855,173 | bf16_mfu: 52.53 | total time: 4.09m | eta: 108.5m step 00213/05568 (3.83%) | loss: 3.409657 | lrm: 1.00 | dt: 1217.39ms | tok/sec: 861,328 | bf16_mfu: 52.91 | total time: 4.11m | eta: 108.5m step 00214/05568 (3.84%) | loss: 3.403201 | lrm: 1.00 | dt: 1215.02ms | tok/sec: 863,008 | bf16_mfu: 53.01 | total time: 4.13m | eta: 108.4m step 00215/05568 (3.86%) | loss: 3.404898 | lrm: 1.00 | dt: 1219.97ms | tok/sec: 859,507 | bf16_mfu: 52.80 | total time: 4.15m | eta: 108.4m step 00216/05568 (3.88%) | loss: 3.401129 | lrm: 1.00 | dt: 1219.49ms | tok/sec: 859,847 | bf16_mfu: 52.82 | total time: 4.17m | eta: 108.4m step 00217/05568 (3.90%) | loss: 3.403004 | lrm: 1.00 | dt: 1217.40ms | tok/sec: 861,325 | bf16_mfu: 52.91 | total time: 4.19m | eta: 108.4m step 00218/05568 (3.92%) | loss: 3.402634 | lrm: 1.00 | dt: 1217.57ms | tok/sec: 861,205 | bf16_mfu: 52.90 | total time: 4.21m | eta: 108.4m step 00219/05568 (3.93%) | loss: 3.408016 | lrm: 1.00 | dt: 1223.41ms | tok/sec: 857,092 | bf16_mfu: 52.65 | total time: 4.23m | eta: 108.3m step 00220/05568 (3.95%) | loss: 3.409566 | lrm: 1.00 | dt: 1218.39ms | tok/sec: 860,625 | bf16_mfu: 52.87 | total time: 4.25m | eta: 108.3m step 00221/05568 (3.97%) | loss: 3.404464 | lrm: 1.00 | dt: 1222.52ms | tok/sec: 857,714 | bf16_mfu: 52.69 | total time: 4.27m | eta: 108.3m step 00222/05568 (3.99%) | loss: 3.402885 | lrm: 1.00 | dt: 1214.16ms | tok/sec: 863,624 | bf16_mfu: 53.05 | total time: 4.30m | eta: 108.3m step 00223/05568 (4.01%) | loss: 3.403170 | lrm: 1.00 | dt: 1209.27ms | tok/sec: 867,117 | bf16_mfu: 53.27 | total time: 4.32m | eta: 108.3m step 00224/05568 (4.02%) | loss: 3.388486 | lrm: 1.00 | dt: 1220.01ms | tok/sec: 859,480 | bf16_mfu: 52.80 | total time: 4.34m | eta: 108.2m step 00225/05568 (4.04%) | loss: 3.368379 | lrm: 1.00 | dt: 1214.02ms | tok/sec: 863,723 | bf16_mfu: 53.06 | total time: 4.36m | eta: 108.2m step 00226/05568 (4.06%) | loss: 3.370044 | lrm: 1.00 | dt: 1218.79ms | tok/sec: 860,340 | bf16_mfu: 52.85 | total time: 4.38m | eta: 108.2m step 00227/05568 (4.08%) | loss: 3.360453 | lrm: 1.00 | dt: 1215.13ms | tok/sec: 862,932 | bf16_mfu: 53.01 | total time: 4.40m | eta: 108.2m step 00228/05568 (4.09%) | loss: 3.360176 | lrm: 1.00 | dt: 1213.30ms | tok/sec: 864,232 | bf16_mfu: 53.09 | total time: 4.42m | eta: 108.2m step 00229/05568 (4.11%) | loss: 3.357866 | lrm: 1.00 | dt: 1214.23ms | tok/sec: 863,572 | bf16_mfu: 53.05 | total time: 4.44m | eta: 108.1m step 00230/05568 (4.13%) | loss: 3.351596 | lrm: 1.00 | dt: 1208.01ms | tok/sec: 868,022 | bf16_mfu: 53.32 | total time: 4.46m | eta: 108.1m step 00231/05568 (4.15%) | loss: 3.342951 | lrm: 1.00 | dt: 1210.63ms | tok/sec: 866,140 | bf16_mfu: 53.21 | total time: 4.48m | eta: 108.1m step 00232/05568 (4.17%) | loss: 3.326349 | lrm: 1.00 | dt: 1213.39ms | tok/sec: 864,167 | bf16_mfu: 53.08 | total time: 4.50m | eta: 108.1m step 00233/05568 (4.18%) | loss: 3.335498 | lrm: 1.00 | dt: 1216.52ms | tok/sec: 861,944 | bf16_mfu: 52.95 | total time: 4.52m | eta: 108.1m step 00234/05568 (4.20%) | loss: 3.327765 | lrm: 1.00 | dt: 1213.93ms | tok/sec: 863,782 | bf16_mfu: 53.06 | total time: 4.54m | eta: 108.0m step 00235/05568 (4.22%) | loss: 3.323704 | lrm: 1.00 | dt: 1216.63ms | tok/sec: 861,869 | bf16_mfu: 52.94 | total time: 4.56m | eta: 108.0m step 00236/05568 (4.24%) | loss: 3.329648 | lrm: 1.00 | dt: 1212.08ms | tok/sec: 865,104 | bf16_mfu: 53.14 | total time: 4.58m | eta: 108.0m step 00237/05568 (4.26%) | loss: 3.336878 | lrm: 1.00 | dt: 1214.73ms | tok/sec: 863,214 | bf16_mfu: 53.03 | total time: 4.60m | eta: 108.0m step 00238/05568 (4.27%) | loss: 3.323568 | lrm: 1.00 | dt: 1218.45ms | tok/sec: 860,578 | bf16_mfu: 52.86 | total time: 4.62m | eta: 108.0m step 00239/05568 (4.29%) | loss: 3.322560 | lrm: 1.00 | dt: 1216.70ms | tok/sec: 861,820 | bf16_mfu: 52.94 | total time: 4.64m | eta: 107.9m step 00240/05568 (4.31%) | loss: 3.312113 | lrm: 1.00 | dt: 1220.86ms | tok/sec: 858,885 | bf16_mfu: 52.76 | total time: 4.66m | eta: 107.9m step 00241/05568 (4.33%) | loss: 3.306774 | lrm: 1.00 | dt: 1216.46ms | tok/sec: 861,992 | bf16_mfu: 52.95 | total time: 4.68m | eta: 107.9m step 00242/05568 (4.35%) | loss: 3.309472 | lrm: 1.00 | dt: 1221.17ms | tok/sec: 858,664 | bf16_mfu: 52.75 | total time: 4.70m | eta: 107.9m step 00243/05568 (4.36%) | loss: 3.291742 | lrm: 1.00 | dt: 1214.72ms | tok/sec: 863,223 | bf16_mfu: 53.03 | total time: 4.72m | eta: 107.9m step 00244/05568 (4.38%) | loss: 3.287886 | lrm: 1.00 | dt: 1214.75ms | tok/sec: 863,200 | bf16_mfu: 53.02 | total time: 4.74m | eta: 107.8m step 00245/05568 (4.40%) | loss: 3.293825 | lrm: 1.00 | dt: 1213.69ms | tok/sec: 863,958 | bf16_mfu: 53.07 | total time: 4.76m | eta: 107.8m step 00246/05568 (4.42%) | loss: 3.295579 | lrm: 1.00 | dt: 1218.74ms | tok/sec: 860,374 | bf16_mfu: 52.85 | total time: 4.78m | eta: 107.8m step 00247/05568 (4.44%) | loss: 3.304866 | lrm: 1.00 | dt: 1211.60ms | tok/sec: 865,450 | bf16_mfu: 53.16 | total time: 4.80m | eta: 107.8m step 00248/05568 (4.45%) | loss: 3.315185 | lrm: 1.00 | dt: 1214.78ms | tok/sec: 863,185 | bf16_mfu: 53.02 | total time: 4.82m | eta: 107.8m step 00249/05568 (4.47%) | loss: 3.324459 | lrm: 1.00 | dt: 1211.84ms | tok/sec: 865,275 | bf16_mfu: 53.15 | total time: 4.84m | eta: 107.7m step:250/5568 val_bpb:0.997472 val_time:3.95s step 00250/05568 (4.49%) | loss: 3.316437 | lrm: 1.00 | dt: 1215.97ms | tok/sec: 862,335 | bf16_mfu: 52.97 | total time: 4.86m | eta: 107.7m step 00251/05568 (4.51%) | loss: 3.312965 | lrm: 1.00 | dt: 1212.12ms | tok/sec: 865,076 | bf16_mfu: 53.14 | total time: 4.88m | eta: 107.7m step 00252/05568 (4.53%) | loss: 3.315539 | lrm: 1.00 | dt: 1210.20ms | tok/sec: 866,445 | bf16_mfu: 53.22 | total time: 4.90m | eta: 107.7m step 00253/05568 (4.54%) | loss: 3.320341 | lrm: 1.00 | dt: 1209.03ms | tok/sec: 867,287 | bf16_mfu: 53.28 | total time: 4.92m | eta: 107.7m step 00254/05568 (4.56%) | loss: 3.305987 | lrm: 1.00 | dt: 1208.94ms | tok/sec: 867,348 | bf16_mfu: 53.28 | total time: 4.94m | eta: 107.6m step 00255/05568 (4.58%) | loss: 3.312674 | lrm: 1.00 | dt: 1215.04ms | tok/sec: 862,999 | bf16_mfu: 53.01 | total time: 4.96m | eta: 107.6m step 00256/05568 (4.60%) | loss: 3.311741 | lrm: 1.00 | dt: 1214.40ms | tok/sec: 863,455 | bf16_mfu: 53.04 | total time: 4.98m | eta: 107.6m step 00257/05568 (4.62%) | loss: 3.306645 | lrm: 1.00 | dt: 1215.40ms | tok/sec: 862,738 | bf16_mfu: 53.00 | total time: 5.00m | eta: 107.6m step 00258/05568 (4.63%) | loss: 3.305347 | lrm: 1.00 | dt: 1217.53ms | tok/sec: 861,235 | bf16_mfu: 52.90 | total time: 5.02m | eta: 107.5m step 00259/05568 (4.65%) | loss: 3.311929 | lrm: 1.00 | dt: 1207.05ms | tok/sec: 868,710 | bf16_mfu: 53.36 | total time: 5.04m | eta: 107.5m step 00260/05568 (4.67%) | loss: 3.290899 | lrm: 1.00 | dt: 1212.52ms | tok/sec: 864,790 | bf16_mfu: 53.12 | total time: 5.06m | eta: 107.5m step 00261/05568 (4.69%) | loss: 3.293022 | lrm: 1.00 | dt: 1216.24ms | tok/sec: 862,142 | bf16_mfu: 52.96 | total time: 5.08m | eta: 107.5m step 00262/05568 (4.71%) | loss: 3.285852 | lrm: 1.00 | dt: 1215.04ms | tok/sec: 862,996 | bf16_mfu: 53.01 | total time: 5.10m | eta: 107.5m step 00263/05568 (4.72%) | loss: 3.273762 | lrm: 1.00 | dt: 1217.97ms | tok/sec: 860,922 | bf16_mfu: 52.88 | total time: 5.13m | eta: 107.4m step 00264/05568 (4.74%) | loss: 3.268891 | lrm: 1.00 | dt: 1210.92ms | tok/sec: 865,934 | bf16_mfu: 53.19 | total time: 5.15m | eta: 107.4m step 00265/05568 (4.76%) | loss: 3.255756 | lrm: 1.00 | dt: 1219.85ms | tok/sec: 859,596 | bf16_mfu: 52.80 | total time: 5.17m | eta: 107.4m step 00266/05568 (4.78%) | loss: 3.257960 | lrm: 1.00 | dt: 1207.75ms | tok/sec: 868,204 | bf16_mfu: 53.33 | total time: 5.19m | eta: 107.4m step 00267/05568 (4.80%) | loss: 3.267055 | lrm: 1.00 | dt: 1212.38ms | tok/sec: 864,890 | bf16_mfu: 53.13 | total time: 5.21m | eta: 107.4m step 00268/05568 (4.81%) | loss: 3.263812 | lrm: 1.00 | dt: 1210.30ms | tok/sec: 866,376 | bf16_mfu: 53.22 | total time: 5.23m | eta: 107.3m step 00269/05568 (4.83%) | loss: 3.253663 | lrm: 1.00 | dt: 1208.41ms | tok/sec: 867,730 | bf16_mfu: 53.30 | total time: 5.25m | eta: 107.3m step 00270/05568 (4.85%) | loss: 3.252917 | lrm: 1.00 | dt: 1211.02ms | tok/sec: 865,859 | bf16_mfu: 53.19 | total time: 5.27m | eta: 107.3m step 00271/05568 (4.87%) | loss: 3.259515 | lrm: 1.00 | dt: 1209.09ms | tok/sec: 867,244 | bf16_mfu: 53.27 | total time: 5.29m | eta: 107.3m step 00272/05568 (4.89%) | loss: 3.257808 | lrm: 1.00 | dt: 1213.96ms | tok/sec: 863,764 | bf16_mfu: 53.06 | total time: 5.31m | eta: 107.2m step 00273/05568 (4.90%) | loss: 3.258509 | lrm: 1.00 | dt: 1214.55ms | tok/sec: 863,346 | bf16_mfu: 53.03 | total time: 5.33m | eta: 107.2m step 00274/05568 (4.92%) | loss: 3.261438 | lrm: 1.00 | dt: 1211.84ms | tok/sec: 865,272 | bf16_mfu: 53.15 | total time: 5.35m | eta: 107.2m step 00275/05568 (4.94%) | loss: 3.261763 | lrm: 1.00 | dt: 1210.30ms | tok/sec: 866,378 | bf16_mfu: 53.22 | total time: 5.37m | eta: 107.2m step 00276/05568 (4.96%) | loss: 3.272703 | lrm: 1.00 | dt: 1215.99ms | tok/sec: 862,323 | bf16_mfu: 52.97 | total time: 5.39m | eta: 107.2m step 00277/05568 (4.97%) | loss: 3.266738 | lrm: 1.00 | dt: 1216.71ms | tok/sec: 861,814 | bf16_mfu: 52.94 | total time: 5.41m | eta: 107.1m step 00278/05568 (4.99%) | loss: 3.251708 | lrm: 1.00 | dt: 1205.86ms | tok/sec: 869,570 | bf16_mfu: 53.42 | total time: 5.43m | eta: 107.1m step 00279/05568 (5.01%) | loss: 3.238792 | lrm: 1.00 | dt: 1212.47ms | tok/sec: 864,826 | bf16_mfu: 53.12 | total time: 5.45m | eta: 107.1m step 00280/05568 (5.03%) | loss: 3.232502 | lrm: 1.00 | dt: 1212.57ms | tok/sec: 864,753 | bf16_mfu: 53.12 | total time: 5.47m | eta: 107.1m step 00281/05568 (5.05%) | loss: 3.226471 | lrm: 1.00 | dt: 1211.05ms | tok/sec: 865,838 | bf16_mfu: 53.19 | total time: 5.49m | eta: 107.1m step 00282/05568 (5.06%) | loss: 3.224530 | lrm: 1.00 | dt: 1209.04ms | tok/sec: 867,277 | bf16_mfu: 53.27 | total time: 5.51m | eta: 107.0m step 00283/05568 (5.08%) | loss: 3.223558 | lrm: 1.00 | dt: 1209.03ms | tok/sec: 867,288 | bf16_mfu: 53.28 | total time: 5.53m | eta: 107.0m step 00284/05568 (5.10%) | loss: 3.230027 | lrm: 1.00 | dt: 1206.73ms | tok/sec: 868,943 | bf16_mfu: 53.38 | total time: 5.55m | eta: 107.0m step 00285/05568 (5.12%) | loss: 3.226431 | lrm: 1.00 | dt: 1208.14ms | tok/sec: 867,926 | bf16_mfu: 53.31 | total time: 5.57m | eta: 107.0m step 00286/05568 (5.14%) | loss: 3.226483 | lrm: 1.00 | dt: 1213.38ms | tok/sec: 864,179 | bf16_mfu: 53.08 | total time: 5.59m | eta: 106.9m step 00287/05568 (5.15%) | loss: 3.227945 | lrm: 1.00 | dt: 1207.80ms | tok/sec: 868,171 | bf16_mfu: 53.33 | total time: 5.61m | eta: 106.9m step 00288/05568 (5.17%) | loss: 3.220560 | lrm: 1.00 | dt: 1206.36ms | tok/sec: 869,208 | bf16_mfu: 53.39 | total time: 5.63m | eta: 106.9m step 00289/05568 (5.19%) | loss: 3.215582 | lrm: 1.00 | dt: 1210.86ms | tok/sec: 865,973 | bf16_mfu: 53.19 | total time: 5.65m | eta: 106.9m step 00290/05568 (5.21%) | loss: 3.223543 | lrm: 1.00 | dt: 1211.75ms | tok/sec: 865,338 | bf16_mfu: 53.16 | total time: 5.67m | eta: 106.9m step 00291/05568 (5.23%) | loss: 3.222917 | lrm: 1.00 | dt: 1210.33ms | tok/sec: 866,357 | bf16_mfu: 53.22 | total time: 5.69m | eta: 106.8m step 00292/05568 (5.24%) | loss: 3.213653 | lrm: 1.00 | dt: 1209.36ms | tok/sec: 867,053 | bf16_mfu: 53.26 | total time: 5.71m | eta: 106.8m step 00293/05568 (5.26%) | loss: 3.208431 | lrm: 1.00 | dt: 1212.33ms | tok/sec: 864,922 | bf16_mfu: 53.13 | total time: 5.73m | eta: 106.8m step 00294/05568 (5.28%) | loss: 3.209817 | lrm: 1.00 | dt: 1209.73ms | tok/sec: 866,784 | bf16_mfu: 53.24 | total time: 5.75m | eta: 106.8m step 00295/05568 (5.30%) | loss: 3.199356 | lrm: 1.00 | dt: 1217.32ms | tok/sec: 861,380 | bf16_mfu: 52.91 | total time: 5.77m | eta: 106.8m step 00296/05568 (5.32%) | loss: 3.200949 | lrm: 1.00 | dt: 1211.16ms | tok/sec: 865,759 | bf16_mfu: 53.18 | total time: 5.79m | eta: 106.7m step 00297/05568 (5.33%) | loss: 3.205764 | lrm: 1.00 | dt: 1203.26ms | tok/sec: 871,447 | bf16_mfu: 53.53 | total time: 5.81m | eta: 106.7m step 00298/05568 (5.35%) | loss: 3.213442 | lrm: 1.00 | dt: 1213.38ms | tok/sec: 864,181 | bf16_mfu: 53.08 | total time: 5.83m | eta: 106.7m step 00299/05568 (5.37%) | loss: 3.198939 | lrm: 1.00 | dt: 1207.71ms | tok/sec: 868,233 | bf16_mfu: 53.33 | total time: 5.85m | eta: 106.7m step 00300/05568 (5.39%) | loss: 3.194762 | lrm: 1.00 | dt: 1211.94ms | tok/sec: 865,203 | bf16_mfu: 53.15 | total time: 5.87m | eta: 106.6m step 00301/05568 (5.41%) | loss: 3.196232 | lrm: 1.00 | dt: 1208.35ms | tok/sec: 867,777 | bf16_mfu: 53.31 | total time: 5.89m | eta: 106.6m step 00302/05568 (5.42%) | loss: 3.199346 | lrm: 1.00 | dt: 1209.76ms | tok/sec: 866,766 | bf16_mfu: 53.24 | total time: 5.91m | eta: 106.6m step 00303/05568 (5.44%) | loss: 3.193923 | lrm: 1.00 | dt: 1206.93ms | tok/sec: 868,797 | bf16_mfu: 53.37 | total time: 5.93m | eta: 106.6m step 00304/05568 (5.46%) | loss: 3.188379 | lrm: 1.00 | dt: 1210.35ms | tok/sec: 866,342 | bf16_mfu: 53.22 | total time: 5.95m | eta: 106.6m step 00305/05568 (5.48%) | loss: 3.184462 | lrm: 1.00 | dt: 1206.05ms | tok/sec: 869,426 | bf16_mfu: 53.41 | total time: 5.97m | eta: 106.5m step 00306/05568 (5.50%) | loss: 3.187193 | lrm: 1.00 | dt: 1207.14ms | tok/sec: 868,644 | bf16_mfu: 53.36 | total time: 5.99m | eta: 106.5m step 00307/05568 (5.51%) | loss: 3.189404 | lrm: 1.00 | dt: 1212.94ms | tok/sec: 864,493 | bf16_mfu: 53.10 | total time: 6.01m | eta: 106.5m step 00308/05568 (5.53%) | loss: 3.174685 | lrm: 1.00 | dt: 1205.60ms | tok/sec: 869,753 | bf16_mfu: 53.43 | total time: 6.03m | eta: 106.5m step 00309/05568 (5.55%) | loss: 3.175270 | lrm: 1.00 | dt: 1220.05ms | tok/sec: 859,454 | bf16_mfu: 52.79 | total time: 6.05m | eta: 106.4m step 00310/05568 (5.57%) | loss: 3.173534 | lrm: 1.00 | dt: 1214.15ms | tok/sec: 863,627 | bf16_mfu: 53.05 | total time: 6.07m | eta: 106.4m step 00311/05568 (5.59%) | loss: 3.170138 | lrm: 1.00 | dt: 1211.26ms | tok/sec: 865,687 | bf16_mfu: 53.18 | total time: 6.09m | eta: 106.4m step 00312/05568 (5.60%) | loss: 3.166417 | lrm: 1.00 | dt: 1211.18ms | tok/sec: 865,745 | bf16_mfu: 53.18 | total time: 6.11m | eta: 106.4m step 00313/05568 (5.62%) | loss: 3.160250 | lrm: 1.00 | dt: 1209.75ms | tok/sec: 866,771 | bf16_mfu: 53.24 | total time: 6.13m | eta: 106.4m step 00314/05568 (5.64%) | loss: 3.164232 | lrm: 1.00 | dt: 1209.79ms | tok/sec: 866,741 | bf16_mfu: 53.24 | total time: 6.15m | eta: 106.3m step 00315/05568 (5.66%) | loss: 3.170493 | lrm: 1.00 | dt: 1207.76ms | tok/sec: 868,197 | bf16_mfu: 53.33 | total time: 6.17m | eta: 106.3m step 00316/05568 (5.68%) | loss: 3.177907 | lrm: 1.00 | dt: 1208.81ms | tok/sec: 867,442 | bf16_mfu: 53.28 | total time: 6.19m | eta: 106.3m step 00317/05568 (5.69%) | loss: 3.182630 | lrm: 1.00 | dt: 1210.97ms | tok/sec: 865,895 | bf16_mfu: 53.19 | total time: 6.21m | eta: 106.3m step 00318/05568 (5.71%) | loss: 3.178160 | lrm: 1.00 | dt: 1215.69ms | tok/sec: 862,536 | bf16_mfu: 52.98 | total time: 6.23m | eta: 106.3m step 00319/05568 (5.73%) | loss: 3.180378 | lrm: 1.00 | dt: 1212.96ms | tok/sec: 864,475 | bf16_mfu: 53.10 | total time: 6.26m | eta: 106.2m step 00320/05568 (5.75%) | loss: 3.185567 | lrm: 1.00 | dt: 1209.20ms | tok/sec: 867,164 | bf16_mfu: 53.27 | total time: 6.28m | eta: 106.2m step 00321/05568 (5.77%) | loss: 3.174924 | lrm: 1.00 | dt: 1209.89ms | tok/sec: 866,668 | bf16_mfu: 53.24 | total time: 6.30m | eta: 106.2m step 00322/05568 (5.78%) | loss: 3.170668 | lrm: 1.00 | dt: 1211.28ms | tok/sec: 865,679 | bf16_mfu: 53.18 | total time: 6.32m | eta: 106.2m step 00323/05568 (5.80%) | loss: 3.165418 | lrm: 1.00 | dt: 1210.55ms | tok/sec: 866,198 | bf16_mfu: 53.21 | total time: 6.34m | eta: 106.1m step 00324/05568 (5.82%) | loss: 3.165840 | lrm: 1.00 | dt: 1202.34ms | tok/sec: 872,112 | bf16_mfu: 53.57 | total time: 6.36m | eta: 106.1m step 00325/05568 (5.84%) | loss: 3.169771 | lrm: 1.00 | dt: 1214.60ms | tok/sec: 863,308 | bf16_mfu: 53.03 | total time: 6.38m | eta: 106.1m step 00326/05568 (5.85%) | loss: 3.169319 | lrm: 1.00 | dt: 1211.57ms | tok/sec: 865,472 | bf16_mfu: 53.16 | total time: 6.40m | eta: 106.1m step 00327/05568 (5.87%) | loss: 3.164518 | lrm: 1.00 | dt: 1210.25ms | tok/sec: 866,411 | bf16_mfu: 53.22 | total time: 6.42m | eta: 106.1m step 00328/05568 (5.89%) | loss: 3.163505 | lrm: 1.00 | dt: 1205.87ms | tok/sec: 869,562 | bf16_mfu: 53.42 | total time: 6.44m | eta: 106.0m step 00329/05568 (5.91%) | loss: 3.164098 | lrm: 1.00 | dt: 1206.62ms | tok/sec: 869,016 | bf16_mfu: 53.38 | total time: 6.46m | eta: 106.0m step 00330/05568 (5.93%) | loss: 3.172606 | lrm: 1.00 | dt: 1208.18ms | tok/sec: 867,894 | bf16_mfu: 53.31 | total time: 6.48m | eta: 106.0m step 00331/05568 (5.94%) | loss: 3.169878 | lrm: 1.00 | dt: 1211.01ms | tok/sec: 865,870 | bf16_mfu: 53.19 | total time: 6.50m | eta: 106.0m step 00332/05568 (5.96%) | loss: 3.168254 | lrm: 1.00 | dt: 1205.48ms | tok/sec: 869,842 | bf16_mfu: 53.43 | total time: 6.52m | eta: 106.0m step 00333/05568 (5.98%) | loss: 3.166895 | lrm: 1.00 | dt: 1208.23ms | tok/sec: 867,862 | bf16_mfu: 53.31 | total time: 6.54m | eta: 105.9m step 00334/05568 (6.00%) | loss: 3.159822 | lrm: 1.00 | dt: 1207.76ms | tok/sec: 868,201 | bf16_mfu: 53.33 | total time: 6.56m | eta: 105.9m step 00335/05568 (6.02%) | loss: 3.161610 | lrm: 1.00 | dt: 1208.47ms | tok/sec: 867,689 | bf16_mfu: 53.30 | total time: 6.58m | eta: 105.9m step 00336/05568 (6.03%) | loss: 3.164396 | lrm: 1.00 | dt: 1206.66ms | tok/sec: 868,989 | bf16_mfu: 53.38 | total time: 6.60m | eta: 105.9m step 00337/05568 (6.05%) | loss: 3.154457 | lrm: 1.00 | dt: 1211.14ms | tok/sec: 865,774 | bf16_mfu: 53.18 | total time: 6.62m | eta: 105.8m step 00338/05568 (6.07%) | loss: 3.139972 | lrm: 1.00 | dt: 1207.84ms | tok/sec: 868,138 | bf16_mfu: 53.33 | total time: 6.64m | eta: 105.8m step 00339/05568 (6.09%) | loss: 3.134989 | lrm: 1.00 | dt: 1204.63ms | tok/sec: 870,451 | bf16_mfu: 53.47 | total time: 6.66m | eta: 105.8m step 00340/05568 (6.11%) | loss: 3.121367 | lrm: 1.00 | dt: 1209.46ms | tok/sec: 866,977 | bf16_mfu: 53.26 | total time: 6.68m | eta: 105.8m step 00341/05568 (6.12%) | loss: 3.118619 | lrm: 1.00 | dt: 1207.34ms | tok/sec: 868,502 | bf16_mfu: 53.35 | total time: 6.70m | eta: 105.8m step 00342/05568 (6.14%) | loss: 3.125334 | lrm: 1.00 | dt: 1214.15ms | tok/sec: 863,630 | bf16_mfu: 53.05 | total time: 6.72m | eta: 105.7m step 00343/05568 (6.16%) | loss: 3.117682 | lrm: 1.00 | dt: 1209.41ms | tok/sec: 867,014 | bf16_mfu: 53.26 | total time: 6.74m | eta: 105.7m step 00344/05568 (6.18%) | loss: 3.122983 | lrm: 1.00 | dt: 1212.17ms | tok/sec: 865,038 | bf16_mfu: 53.14 | total time: 6.76m | eta: 105.7m step 00345/05568 (6.20%) | loss: 3.127304 | lrm: 1.00 | dt: 1207.36ms | tok/sec: 868,485 | bf16_mfu: 53.35 | total time: 6.78m | eta: 105.7m step 00346/05568 (6.21%) | loss: 3.122536 | lrm: 1.00 | dt: 1204.01ms | tok/sec: 870,903 | bf16_mfu: 53.50 | total time: 6.80m | eta: 105.6m step 00347/05568 (6.23%) | loss: 3.122276 | lrm: 1.00 | dt: 1204.78ms | tok/sec: 870,349 | bf16_mfu: 53.46 | total time: 6.82m | eta: 105.6m step 00348/05568 (6.25%) | loss: 3.131646 | lrm: 1.00 | dt: 1208.81ms | tok/sec: 867,443 | bf16_mfu: 53.29 | total time: 6.84m | eta: 105.6m step 00349/05568 (6.27%) | loss: 3.134549 | lrm: 1.00 | dt: 1211.17ms | tok/sec: 865,753 | bf16_mfu: 53.18 | total time: 6.86m | eta: 105.6m step 00350/05568 (6.29%) | loss: 3.125707 | lrm: 1.00 | dt: 1211.96ms | tok/sec: 865,192 | bf16_mfu: 53.15 | total time: 6.88m | eta: 105.6m step 00351/05568 (6.30%) | loss: 3.110999 | lrm: 1.00 | dt: 1204.59ms | tok/sec: 870,481 | bf16_mfu: 53.47 | total time: 6.90m | eta: 105.5m step 00352/05568 (6.32%) | loss: 3.116836 | lrm: 1.00 | dt: 1201.62ms | tok/sec: 872,636 | bf16_mfu: 53.60 | total time: 6.92m | eta: 105.5m step 00353/05568 (6.34%) | loss: 3.113435 | lrm: 1.00 | dt: 1203.44ms | tok/sec: 871,316 | bf16_mfu: 53.52 | total time: 6.94m | eta: 105.5m step 00354/05568 (6.36%) | loss: 3.116755 | lrm: 1.00 | dt: 1205.35ms | tok/sec: 869,934 | bf16_mfu: 53.44 | total time: 6.96m | eta: 105.5m step 00355/05568 (6.38%) | loss: 3.102818 | lrm: 1.00 | dt: 1209.33ms | tok/sec: 867,073 | bf16_mfu: 53.26 | total time: 6.98m | eta: 105.4m step 00356/05568 (6.39%) | loss: 3.097854 | lrm: 1.00 | dt: 1210.40ms | tok/sec: 866,305 | bf16_mfu: 53.22 | total time: 7.00m | eta: 105.4m step 00357/05568 (6.41%) | loss: 3.092385 | lrm: 1.00 | dt: 1203.64ms | tok/sec: 871,169 | bf16_mfu: 53.51 | total time: 7.02m | eta: 105.4m step 00358/05568 (6.43%) | loss: 3.079092 | lrm: 1.00 | dt: 1211.76ms | tok/sec: 865,336 | bf16_mfu: 53.16 | total time: 7.04m | eta: 105.4m step 00359/05568 (6.45%) | loss: 3.086334 | lrm: 1.00 | dt: 1209.23ms | tok/sec: 867,140 | bf16_mfu: 53.27 | total time: 7.06m | eta: 105.4m step 00360/05568 (6.47%) | loss: 3.094779 | lrm: 1.00 | dt: 1211.06ms | tok/sec: 865,832 | bf16_mfu: 53.19 | total time: 7.08m | eta: 105.3m step 00361/05568 (6.48%) | loss: 3.092173 | lrm: 1.00 | dt: 1204.80ms | tok/sec: 870,328 | bf16_mfu: 53.46 | total time: 7.10m | eta: 105.3m step 00362/05568 (6.50%) | loss: 3.090290 | lrm: 1.00 | dt: 1202.99ms | tok/sec: 871,644 | bf16_mfu: 53.54 | total time: 7.12m | eta: 105.3m step 00363/05568 (6.52%) | loss: 3.086364 | lrm: 1.00 | dt: 1205.82ms | tok/sec: 869,592 | bf16_mfu: 53.42 | total time: 7.14m | eta: 105.3m step 00364/05568 (6.54%) | loss: 3.093931 | lrm: 1.00 | dt: 1208.20ms | tok/sec: 867,883 | bf16_mfu: 53.31 | total time: 7.16m | eta: 105.3m step 00365/05568 (6.56%) | loss: 3.090792 | lrm: 1.00 | dt: 1202.04ms | tok/sec: 872,329 | bf16_mfu: 53.59 | total time: 7.18m | eta: 105.2m step 00366/05568 (6.57%) | loss: 3.084878 | lrm: 1.00 | dt: 1214.47ms | tok/sec: 863,400 | bf16_mfu: 53.04 | total time: 7.20m | eta: 105.2m step 00367/05568 (6.59%) | loss: 3.088027 | lrm: 1.00 | dt: 1206.76ms | tok/sec: 868,920 | bf16_mfu: 53.38 | total time: 7.22m | eta: 105.2m step 00368/05568 (6.61%) | loss: 3.091435 | lrm: 1.00 | dt: 1203.76ms | tok/sec: 871,084 | bf16_mfu: 53.51 | total time: 7.24m | eta: 105.2m step 00369/05568 (6.63%) | loss: 3.093851 | lrm: 1.00 | dt: 1208.50ms | tok/sec: 867,664 | bf16_mfu: 53.30 | total time: 7.26m | eta: 105.1m step 00370/05568 (6.65%) | loss: 3.084314 | lrm: 1.00 | dt: 1210.93ms | tok/sec: 865,928 | bf16_mfu: 53.19 | total time: 7.28m | eta: 105.1m step 00371/05568 (6.66%) | loss: 3.084937 | lrm: 1.00 | dt: 1202.57ms | tok/sec: 871,944 | bf16_mfu: 53.56 | total time: 7.30m | eta: 105.1m step 00372/05568 (6.68%) | loss: 3.089369 | lrm: 1.00 | dt: 1203.73ms | tok/sec: 871,105 | bf16_mfu: 53.51 | total time: 7.32m | eta: 105.1m step 00373/05568 (6.70%) | loss: 3.089698 | lrm: 1.00 | dt: 1202.70ms | tok/sec: 871,854 | bf16_mfu: 53.56 | total time: 7.34m | eta: 105.1m step 00374/05568 (6.72%) | loss: 3.079546 | lrm: 1.00 | dt: 1207.39ms | tok/sec: 868,463 | bf16_mfu: 53.35 | total time: 7.36m | eta: 105.0m step 00375/05568 (6.73%) | loss: 3.086958 | lrm: 1.00 | dt: 1197.69ms | tok/sec: 875,495 | bf16_mfu: 53.78 | total time: 7.38m | eta: 105.0m step 00376/05568 (6.75%) | loss: 3.092540 | lrm: 1.00 | dt: 1205.20ms | tok/sec: 870,043 | bf16_mfu: 53.44 | total time: 7.40m | eta: 105.0m step 00377/05568 (6.77%) | loss: 3.088607 | lrm: 1.00 | dt: 1206.11ms | tok/sec: 869,383 | bf16_mfu: 53.40 | total time: 7.42m | eta: 105.0m step 00378/05568 (6.79%) | loss: 3.085647 | lrm: 1.00 | dt: 1200.96ms | tok/sec: 873,115 | bf16_mfu: 53.63 | total time: 7.44m | eta: 104.9m step 00379/05568 (6.81%) | loss: 3.088984 | lrm: 1.00 | dt: 1207.08ms | tok/sec: 868,685 | bf16_mfu: 53.36 | total time: 7.46m | eta: 104.9m step 00380/05568 (6.82%) | loss: 3.086358 | lrm: 1.00 | dt: 1199.03ms | tok/sec: 874,522 | bf16_mfu: 53.72 | total time: 7.48m | eta: 104.9m step 00381/05568 (6.84%) | loss: 3.082302 | lrm: 1.00 | dt: 1204.94ms | tok/sec: 870,229 | bf16_mfu: 53.46 | total time: 7.50m | eta: 104.9m step 00382/05568 (6.86%) | loss: 3.086855 | lrm: 1.00 | dt: 1202.08ms | tok/sec: 872,298 | bf16_mfu: 53.58 | total time: 7.52m | eta: 104.9m step 00383/05568 (6.88%) | loss: 3.088337 | lrm: 1.00 | dt: 1203.47ms | tok/sec: 871,292 | bf16_mfu: 53.52 | total time: 7.54m | eta: 104.8m step 00384/05568 (6.90%) | loss: 3.088846 | lrm: 1.00 | dt: 1205.92ms | tok/sec: 869,523 | bf16_mfu: 53.41 | total time: 7.56m | eta: 104.8m step 00385/05568 (6.91%) | loss: 3.094391 | lrm: 1.00 | dt: 1205.59ms | tok/sec: 869,764 | bf16_mfu: 53.43 | total time: 7.58m | eta: 104.8m step 00386/05568 (6.93%) | loss: 3.090646 | lrm: 1.00 | dt: 1201.59ms | tok/sec: 872,657 | bf16_mfu: 53.61 | total time: 7.60m | eta: 104.8m step 00387/05568 (6.95%) | loss: 3.092127 | lrm: 1.00 | dt: 1212.10ms | tok/sec: 865,088 | bf16_mfu: 53.14 | total time: 7.62m | eta: 104.7m step 00388/05568 (6.97%) | loss: 3.089262 | lrm: 1.00 | dt: 1212.44ms | tok/sec: 864,844 | bf16_mfu: 53.13 | total time: 7.64m | eta: 104.7m step 00389/05568 (6.99%) | loss: 3.096519 | lrm: 1.00 | dt: 1205.19ms | tok/sec: 870,051 | bf16_mfu: 53.45 | total time: 7.66m | eta: 104.7m step 00390/05568 (7.00%) | loss: 3.084232 | lrm: 1.00 | dt: 1206.89ms | tok/sec: 868,827 | bf16_mfu: 53.37 | total time: 7.68m | eta: 104.7m step 00391/05568 (7.02%) | loss: 3.082999 | lrm: 1.00 | dt: 1200.76ms | tok/sec: 873,259 | bf16_mfu: 53.64 | total time: 7.70m | eta: 104.7m step 00392/05568 (7.04%) | loss: 3.080533 | lrm: 1.00 | dt: 1206.40ms | tok/sec: 869,179 | bf16_mfu: 53.39 | total time: 7.72m | eta: 104.6m step 00393/05568 (7.06%) | loss: 3.083848 | lrm: 1.00 | dt: 1204.38ms | tok/sec: 870,637 | bf16_mfu: 53.48 | total time: 7.74m | eta: 104.6m step 00394/05568 (7.08%) | loss: 3.080122 | lrm: 1.00 | dt: 1200.70ms | tok/sec: 873,304 | bf16_mfu: 53.65 | total time: 7.76m | eta: 104.6m step 00395/05568 (7.09%) | loss: 3.072182 | lrm: 1.00 | dt: 1203.84ms | tok/sec: 871,022 | bf16_mfu: 53.50 | total time: 7.78m | eta: 104.6m step 00396/05568 (7.11%) | loss: 3.073955 | lrm: 1.00 | dt: 1205.24ms | tok/sec: 870,014 | bf16_mfu: 53.44 | total time: 7.80m | eta: 104.5m step 00397/05568 (7.13%) | loss: 3.078849 | lrm: 1.00 | dt: 1205.82ms | tok/sec: 869,593 | bf16_mfu: 53.42 | total time: 7.82m | eta: 104.5m step 00398/05568 (7.15%) | loss: 3.091214 | lrm: 1.00 | dt: 1208.07ms | tok/sec: 867,978 | bf16_mfu: 53.32 | total time: 7.84m | eta: 104.5m step 00399/05568 (7.17%) | loss: 3.097523 | lrm: 1.00 | dt: 1203.79ms | tok/sec: 871,064 | bf16_mfu: 53.51 | total time: 7.86m | eta: 104.5m step 00400/05568 (7.18%) | loss: 3.094492 | lrm: 1.00 | dt: 1204.66ms | tok/sec: 870,435 | bf16_mfu: 53.47 | total time: 7.88m | eta: 104.5m step 00401/05568 (7.20%) | loss: 3.094755 | lrm: 1.00 | dt: 1203.73ms | tok/sec: 871,108 | bf16_mfu: 53.51 | total time: 7.90m | eta: 104.4m step 00402/05568 (7.22%) | loss: 3.087907 | lrm: 1.00 | dt: 1200.69ms | tok/sec: 873,310 | bf16_mfu: 53.65 | total time: 7.92m | eta: 104.4m step 00403/05568 (7.24%) | loss: 3.078476 | lrm: 1.00 | dt: 1205.53ms | tok/sec: 869,803 | bf16_mfu: 53.43 | total time: 7.94m | eta: 104.4m step 00404/05568 (7.26%) | loss: 3.079880 | lrm: 1.00 | dt: 1201.65ms | tok/sec: 872,610 | bf16_mfu: 53.60 | total time: 7.96m | eta: 104.4m step 00405/05568 (7.27%) | loss: 3.077502 | lrm: 1.00 | dt: 1201.34ms | tok/sec: 872,838 | bf16_mfu: 53.62 | total time: 7.98m | eta: 104.3m step 00406/05568 (7.29%) | loss: 3.079528 | lrm: 1.00 | dt: 1202.19ms | tok/sec: 872,218 | bf16_mfu: 53.58 | total time: 8.00m | eta: 104.3m step 00407/05568 (7.31%) | loss: 3.065640 | lrm: 1.00 | dt: 1201.67ms | tok/sec: 872,599 | bf16_mfu: 53.60 | total time: 8.02m | eta: 104.3m step 00408/05568 (7.33%) | loss: 3.063386 | lrm: 1.00 | dt: 1202.27ms | tok/sec: 872,164 | bf16_mfu: 53.58 | total time: 8.04m | eta: 104.3m step 00409/05568 (7.35%) | loss: 3.068890 | lrm: 1.00 | dt: 1204.76ms | tok/sec: 870,359 | bf16_mfu: 53.46 | total time: 8.06m | eta: 104.3m step 00410/05568 (7.36%) | loss: 3.065810 | lrm: 1.00 | dt: 1208.19ms | tok/sec: 867,890 | bf16_mfu: 53.31 | total time: 8.08m | eta: 104.2m step 00411/05568 (7.38%) | loss: 3.051944 | lrm: 1.00 | dt: 1201.97ms | tok/sec: 872,383 | bf16_mfu: 53.59 | total time: 8.10m | eta: 104.2m step 00412/05568 (7.40%) | loss: 3.055997 | lrm: 1.00 | dt: 1198.38ms | tok/sec: 874,997 | bf16_mfu: 53.75 | total time: 8.12m | eta: 104.2m step 00413/05568 (7.42%) | loss: 3.060121 | lrm: 1.00 | dt: 1199.23ms | tok/sec: 874,376 | bf16_mfu: 53.71 | total time: 8.14m | eta: 104.2m step 00414/05568 (7.44%) | loss: 3.065169 | lrm: 1.00 | dt: 1204.99ms | tok/sec: 870,196 | bf16_mfu: 53.45 | total time: 8.16m | eta: 104.1m step 00415/05568 (7.45%) | loss: 3.066593 | lrm: 1.00 | dt: 1204.00ms | tok/sec: 870,909 | bf16_mfu: 53.50 | total time: 8.18m | eta: 104.1m step 00416/05568 (7.47%) | loss: 3.073974 | lrm: 1.00 | dt: 1207.14ms | tok/sec: 868,641 | bf16_mfu: 53.36 | total time: 8.21m | eta: 104.1m step 00417/05568 (7.49%) | loss: 3.072179 | lrm: 1.00 | dt: 1201.89ms | tok/sec: 872,439 | bf16_mfu: 53.59 | total time: 8.23m | eta: 104.1m step 00418/05568 (7.51%) | loss: 3.060478 | lrm: 1.00 | dt: 1197.18ms | tok/sec: 875,869 | bf16_mfu: 53.80 | total time: 8.25m | eta: 104.1m step 00419/05568 (7.53%) | loss: 3.063438 | lrm: 1.00 | dt: 1205.41ms | tok/sec: 869,893 | bf16_mfu: 53.44 | total time: 8.27m | eta: 104.0m step 00420/05568 (7.54%) | loss: 3.071957 | lrm: 1.00 | dt: 1202.91ms | tok/sec: 871,698 | bf16_mfu: 53.55 | total time: 8.29m | eta: 104.0m step 00421/05568 (7.56%) | loss: 3.071817 | lrm: 1.00 | dt: 1201.48ms | tok/sec: 872,733 | bf16_mfu: 53.61 | total time: 8.31m | eta: 104.0m step 00422/05568 (7.58%) | loss: 3.053542 | lrm: 1.00 | dt: 1202.94ms | tok/sec: 871,674 | bf16_mfu: 53.54 | total time: 8.33m | eta: 104.0m step 00423/05568 (7.60%) | loss: 3.057704 | lrm: 1.00 | dt: 1201.20ms | tok/sec: 872,943 | bf16_mfu: 53.62 | total time: 8.35m | eta: 103.9m step 00424/05568 (7.61%) | loss: 3.068427 | lrm: 1.00 | dt: 1196.70ms | tok/sec: 876,222 | bf16_mfu: 53.82 | total time: 8.37m | eta: 103.9m step 00425/05568 (7.63%) | loss: 3.075014 | lrm: 1.00 | dt: 1194.44ms | tok/sec: 877,879 | bf16_mfu: 53.93 | total time: 8.39m | eta: 103.9m step 00426/05568 (7.65%) | loss: 3.076661 | lrm: 1.00 | dt: 1205.86ms | tok/sec: 869,567 | bf16_mfu: 53.42 | total time: 8.41m | eta: 103.9m step 00427/05568 (7.67%) | loss: 3.061395 | lrm: 1.00 | dt: 1196.13ms | tok/sec: 876,639 | bf16_mfu: 53.85 | total time: 8.43m | eta: 103.8m step 00428/05568 (7.69%) | loss: 3.073375 | lrm: 1.00 | dt: 1195.93ms | tok/sec: 876,786 | bf16_mfu: 53.86 | total time: 8.45m | eta: 103.8m step 00429/05568 (7.70%) | loss: 3.076271 | lrm: 1.00 | dt: 1202.78ms | tok/sec: 871,794 | bf16_mfu: 53.55 | total time: 8.47m | eta: 103.8m step 00430/05568 (7.72%) | loss: 3.076498 | lrm: 1.00 | dt: 1195.71ms | tok/sec: 876,947 | bf16_mfu: 53.87 | total time: 8.49m | eta: 103.8m step 00431/05568 (7.74%) | loss: 3.082508 | lrm: 1.00 | dt: 1196.47ms | tok/sec: 876,389 | bf16_mfu: 53.83 | total time: 8.50m | eta: 103.8m step 00432/05568 (7.76%) | loss: 3.091489 | lrm: 1.00 | dt: 1200.13ms | tok/sec: 873,721 | bf16_mfu: 53.67 | total time: 8.52m | eta: 103.7m step 00433/05568 (7.78%) | loss: 3.086360 | lrm: 1.00 | dt: 1201.09ms | tok/sec: 873,018 | bf16_mfu: 53.63 | total time: 8.54m | eta: 103.7m step 00434/05568 (7.79%) | loss: 3.077245 | lrm: 1.00 | dt: 1194.56ms | tok/sec: 877,792 | bf16_mfu: 53.92 | total time: 8.56m | eta: 103.7m step 00435/05568 (7.81%) | loss: 3.079279 | lrm: 1.00 | dt: 1210.52ms | tok/sec: 866,218 | bf16_mfu: 53.21 | total time: 8.59m | eta: 103.7m step 00436/05568 (7.83%) | loss: 3.070347 | lrm: 1.00 | dt: 1198.66ms | tok/sec: 874,788 | bf16_mfu: 53.74 | total time: 8.61m | eta: 103.6m step 00437/05568 (7.85%) | loss: 3.062444 | lrm: 1.00 | dt: 1203.09ms | tok/sec: 871,569 | bf16_mfu: 53.54 | total time: 8.63m | eta: 103.6m step 00438/05568 (7.87%) | loss: 3.060039 | lrm: 1.00 | dt: 1198.84ms | tok/sec: 874,661 | bf16_mfu: 53.73 | total time: 8.65m | eta: 103.6m step 00439/05568 (7.88%) | loss: 3.052321 | lrm: 1.00 | dt: 1198.46ms | tok/sec: 874,938 | bf16_mfu: 53.75 | total time: 8.67m | eta: 103.6m step 00440/05568 (7.90%) | loss: 3.046742 | lrm: 1.00 | dt: 1195.36ms | tok/sec: 877,208 | bf16_mfu: 53.88 | total time: 8.68m | eta: 103.6m step 00441/05568 (7.92%) | loss: 3.041062 | lrm: 1.00 | dt: 1198.64ms | tok/sec: 874,802 | bf16_mfu: 53.74 | total time: 8.70m | eta: 103.5m step 00442/05568 (7.94%) | loss: 3.032724 | lrm: 1.00 | dt: 1201.49ms | tok/sec: 872,726 | bf16_mfu: 53.61 | total time: 8.72m | eta: 103.5m step 00443/05568 (7.96%) | loss: 3.029285 | lrm: 1.00 | dt: 1201.78ms | tok/sec: 872,522 | bf16_mfu: 53.60 | total time: 8.75m | eta: 103.5m step 00444/05568 (7.97%) | loss: 3.025898 | lrm: 1.00 | dt: 1196.41ms | tok/sec: 876,433 | bf16_mfu: 53.84 | total time: 8.76m | eta: 103.5m step 00445/05568 (7.99%) | loss: 3.030249 | lrm: 1.00 | dt: 1199.35ms | tok/sec: 874,288 | bf16_mfu: 53.71 | total time: 8.78m | eta: 103.4m step 00446/05568 (8.01%) | loss: 3.019438 | lrm: 1.00 | dt: 1198.20ms | tok/sec: 875,123 | bf16_mfu: 53.76 | total time: 8.80m | eta: 103.4m step 00447/05568 (8.03%) | loss: 3.015428 | lrm: 1.00 | dt: 1200.65ms | tok/sec: 873,343 | bf16_mfu: 53.65 | total time: 8.82m | eta: 103.4m step 00448/05568 (8.05%) | loss: 3.016097 | lrm: 1.00 | dt: 1197.88ms | tok/sec: 875,361 | bf16_mfu: 53.77 | total time: 8.84m | eta: 103.4m step 00449/05568 (8.06%) | loss: 3.012273 | lrm: 1.00 | dt: 1200.10ms | tok/sec: 873,739 | bf16_mfu: 53.67 | total time: 8.86m | eta: 103.3m step 00450/05568 (8.08%) | loss: 3.013516 | lrm: 1.00 | dt: 1199.60ms | tok/sec: 874,104 | bf16_mfu: 53.69 | total time: 8.88m | eta: 103.3m step 00451/05568 (8.10%) | loss: 3.015331 | lrm: 1.00 | dt: 1199.58ms | tok/sec: 874,117 | bf16_mfu: 53.70 | total time: 8.90m | eta: 103.3m step 00452/05568 (8.12%) | loss: 3.017049 | lrm: 1.00 | dt: 1196.27ms | tok/sec: 876,535 | bf16_mfu: 53.84 | total time: 8.92m | eta: 103.3m step 00453/05568 (8.14%) | loss: 3.016561 | lrm: 1.00 | dt: 1197.86ms | tok/sec: 875,374 | bf16_mfu: 53.77 | total time: 8.94m | eta: 103.3m step 00454/05568 (8.15%) | loss: 3.000532 | lrm: 1.00 | dt: 1201.98ms | tok/sec: 872,372 | bf16_mfu: 53.59 | total time: 8.96m | eta: 103.2m step 00455/05568 (8.17%) | loss: 3.009606 | lrm: 1.00 | dt: 1197.65ms | tok/sec: 875,530 | bf16_mfu: 53.78 | total time: 8.98m | eta: 103.2m step 00456/05568 (8.19%) | loss: 3.010028 | lrm: 1.00 | dt: 1203.53ms | tok/sec: 871,249 | bf16_mfu: 53.52 | total time: 9.00m | eta: 103.2m step 00457/05568 (8.21%) | loss: 3.012974 | lrm: 1.00 | dt: 1203.65ms | tok/sec: 871,163 | bf16_mfu: 53.51 | total time: 9.02m | eta: 103.2m step 00458/05568 (8.23%) | loss: 3.007789 | lrm: 1.00 | dt: 1194.93ms | tok/sec: 877,523 | bf16_mfu: 53.90 | total time: 9.04m | eta: 103.1m step 00459/05568 (8.24%) | loss: 3.003487 | lrm: 1.00 | dt: 1201.93ms | tok/sec: 872,412 | bf16_mfu: 53.59 | total time: 9.06m | eta: 103.1m step 00460/05568 (8.26%) | loss: 3.012578 | lrm: 1.00 | dt: 1209.92ms | tok/sec: 866,650 | bf16_mfu: 53.24 | total time: 9.09m | eta: 103.1m step 00461/05568 (8.28%) | loss: 3.014456 | lrm: 1.00 | dt: 1201.72ms | tok/sec: 872,562 | bf16_mfu: 53.60 | total time: 9.11m | eta: 103.1m step 00462/05568 (8.30%) | loss: 3.015932 | lrm: 1.00 | dt: 1199.62ms | tok/sec: 874,088 | bf16_mfu: 53.69 | total time: 9.13m | eta: 103.1m step 00463/05568 (8.32%) | loss: 3.020762 | lrm: 1.00 | dt: 1199.31ms | tok/sec: 874,313 | bf16_mfu: 53.71 | total time: 9.15m | eta: 103.0m step 00464/05568 (8.33%) | loss: 3.019164 | lrm: 1.00 | dt: 1201.01ms | tok/sec: 873,078 | bf16_mfu: 53.63 | total time: 9.17m | eta: 103.0m step 00465/05568 (8.35%) | loss: 3.015971 | lrm: 1.00 | dt: 1197.50ms | tok/sec: 875,634 | bf16_mfu: 53.79 | total time: 9.18m | eta: 103.0m step 00466/05568 (8.37%) | loss: 3.022529 | lrm: 1.00 | dt: 1200.09ms | tok/sec: 873,750 | bf16_mfu: 53.67 | total time: 9.20m | eta: 103.0m step 00467/05568 (8.39%) | loss: 3.024781 | lrm: 1.00 | dt: 1194.00ms | tok/sec: 878,202 | bf16_mfu: 53.95 | total time: 9.22m | eta: 102.9m step 00468/05568 (8.41%) | loss: 3.028740 | lrm: 1.00 | dt: 1199.79ms | tok/sec: 873,969 | bf16_mfu: 53.69 | total time: 9.24m | eta: 102.9m step 00469/05568 (8.42%) | loss: 3.030035 | lrm: 1.00 | dt: 1196.84ms | tok/sec: 876,119 | bf16_mfu: 53.82 | total time: 9.26m | eta: 102.9m step 00470/05568 (8.44%) | loss: 3.030226 | lrm: 1.00 | dt: 1199.94ms | tok/sec: 873,859 | bf16_mfu: 53.68 | total time: 9.28m | eta: 102.9m step 00471/05568 (8.46%) | loss: 3.020558 | lrm: 1.00 | dt: 1199.45ms | tok/sec: 874,216 | bf16_mfu: 53.70 | total time: 9.30m | eta: 102.9m step 00472/05568 (8.48%) | loss: 3.018684 | lrm: 1.00 | dt: 1197.09ms | tok/sec: 875,936 | bf16_mfu: 53.81 | total time: 9.32m | eta: 102.8m step 00473/05568 (8.49%) | loss: 3.020030 | lrm: 1.00 | dt: 1200.11ms | tok/sec: 873,730 | bf16_mfu: 53.67 | total time: 9.34m | eta: 102.8m step 00474/05568 (8.51%) | loss: 3.019491 | lrm: 1.00 | dt: 1205.71ms | tok/sec: 869,677 | bf16_mfu: 53.42 | total time: 9.36m | eta: 102.8m step 00475/05568 (8.53%) | loss: 3.018057 | lrm: 1.00 | dt: 1196.66ms | tok/sec: 876,252 | bf16_mfu: 53.83 | total time: 9.38m | eta: 102.8m step 00476/05568 (8.55%) | loss: 3.021067 | lrm: 1.00 | dt: 1196.68ms | tok/sec: 876,239 | bf16_mfu: 53.83 | total time: 9.40m | eta: 102.7m step 00477/05568 (8.57%) | loss: 3.012339 | lrm: 1.00 | dt: 1198.79ms | tok/sec: 874,695 | bf16_mfu: 53.73 | total time: 9.42m | eta: 102.7m step 00478/05568 (8.58%) | loss: 3.010986 | lrm: 1.00 | dt: 1197.22ms | tok/sec: 875,845 | bf16_mfu: 53.80 | total time: 9.44m | eta: 102.7m step 00479/05568 (8.60%) | loss: 3.006272 | lrm: 1.00 | dt: 1199.22ms | tok/sec: 874,382 | bf16_mfu: 53.71 | total time: 9.46m | eta: 102.7m step 00480/05568 (8.62%) | loss: 3.012829 | lrm: 1.00 | dt: 1201.33ms | tok/sec: 872,845 | bf16_mfu: 53.62 | total time: 9.48m | eta: 102.7m step 00481/05568 (8.64%) | loss: 3.007835 | lrm: 1.00 | dt: 1198.16ms | tok/sec: 875,158 | bf16_mfu: 53.76 | total time: 9.50m | eta: 102.6m step 00482/05568 (8.66%) | loss: 3.007400 | lrm: 1.00 | dt: 1199.66ms | tok/sec: 874,059 | bf16_mfu: 53.69 | total time: 9.52m | eta: 102.6m step 00483/05568 (8.67%) | loss: 2.995731 | lrm: 1.00 | dt: 1205.57ms | tok/sec: 869,776 | bf16_mfu: 53.43 | total time: 9.54m | eta: 102.6m step 00484/05568 (8.69%) | loss: 2.990756 | lrm: 1.00 | dt: 1202.82ms | tok/sec: 871,763 | bf16_mfu: 53.55 | total time: 9.56m | eta: 102.6m step 00485/05568 (8.71%) | loss: 2.986614 | lrm: 1.00 | dt: 1197.41ms | tok/sec: 875,704 | bf16_mfu: 53.79 | total time: 9.58m | eta: 102.5m step 00486/05568 (8.73%) | loss: 2.993363 | lrm: 1.00 | dt: 1199.72ms | tok/sec: 874,014 | bf16_mfu: 53.69 | total time: 9.60m | eta: 102.5m step 00487/05568 (8.75%) | loss: 2.977012 | lrm: 1.00 | dt: 1198.50ms | tok/sec: 874,905 | bf16_mfu: 53.74 | total time: 9.62m | eta: 102.5m step 00488/05568 (8.76%) | loss: 2.985044 | lrm: 1.00 | dt: 1198.50ms | tok/sec: 874,903 | bf16_mfu: 53.74 | total time: 9.64m | eta: 102.5m step 00489/05568 (8.78%) | loss: 2.975311 | lrm: 1.00 | dt: 1199.81ms | tok/sec: 873,949 | bf16_mfu: 53.68 | total time: 9.66m | eta: 102.5m step 00490/05568 (8.80%) | loss: 2.981235 | lrm: 1.00 | dt: 1200.34ms | tok/sec: 873,569 | bf16_mfu: 53.66 | total time: 9.68m | eta: 102.4m step 00491/05568 (8.82%) | loss: 2.986669 | lrm: 1.00 | dt: 1196.05ms | tok/sec: 876,702 | bf16_mfu: 53.85 | total time: 9.70m | eta: 102.4m step 00492/05568 (8.84%) | loss: 2.988489 | lrm: 1.00 | dt: 1197.87ms | tok/sec: 875,368 | bf16_mfu: 53.77 | total time: 9.72m | eta: 102.4m step 00493/05568 (8.85%) | loss: 2.980815 | lrm: 1.00 | dt: 1197.32ms | tok/sec: 875,767 | bf16_mfu: 53.80 | total time: 9.74m | eta: 102.4m step 00494/05568 (8.87%) | loss: 2.990669 | lrm: 1.00 | dt: 1199.56ms | tok/sec: 874,131 | bf16_mfu: 53.70 | total time: 9.76m | eta: 102.3m step 00495/05568 (8.89%) | loss: 2.984934 | lrm: 1.00 | dt: 1203.00ms | tok/sec: 871,635 | bf16_mfu: 53.54 | total time: 9.78m | eta: 102.3m step 00496/05568 (8.91%) | loss: 2.989912 | lrm: 1.00 | dt: 1194.30ms | tok/sec: 877,986 | bf16_mfu: 53.93 | total time: 9.80m | eta: 102.3m step 00497/05568 (8.93%) | loss: 2.994900 | lrm: 1.00 | dt: 1205.83ms | tok/sec: 869,586 | bf16_mfu: 53.42 | total time: 9.82m | eta: 102.3m step 00498/05568 (8.94%) | loss: 2.999971 | lrm: 1.00 | dt: 1198.14ms | tok/sec: 875,169 | bf16_mfu: 53.76 | total time: 9.84m | eta: 102.3m step 00499/05568 (8.96%) | loss: 3.006502 | lrm: 1.00 | dt: 1199.86ms | tok/sec: 873,914 | bf16_mfu: 53.68 | total time: 9.86m | eta: 102.2m step:500/5568 val_bpb:0.904081 val_time:3.90s step 00500/05568 (8.98%) | loss: 3.006031 | lrm: 1.00 | dt: 1204.34ms | tok/sec: 870,667 | bf16_mfu: 53.48 | total time: 9.88m | eta: 102.2m step 00501/05568 (9.00%) | loss: 3.007453 | lrm: 1.00 | dt: 1201.44ms | tok/sec: 872,763 | bf16_mfu: 53.61 | total time: 9.90m | eta: 102.2m step 00502/05568 (9.02%) | loss: 3.005777 | lrm: 1.00 | dt: 1196.31ms | tok/sec: 876,510 | bf16_mfu: 53.84 | total time: 9.92m | eta: 102.2m step 00503/05568 (9.03%) | loss: 3.006077 | lrm: 1.00 | dt: 1200.26ms | tok/sec: 873,621 | bf16_mfu: 53.66 | total time: 9.94m | eta: 102.1m step 00504/05568 (9.05%) | loss: 2.998038 | lrm: 1.00 | dt: 1200.84ms | tok/sec: 873,199 | bf16_mfu: 53.64 | total time: 9.96m | eta: 102.1m step 00505/05568 (9.07%) | loss: 2.993688 | lrm: 1.00 | dt: 1196.09ms | tok/sec: 876,667 | bf16_mfu: 53.85 | total time: 9.98m | eta: 102.1m step 00506/05568 (9.09%) | loss: 2.992501 | lrm: 1.00 | dt: 1192.72ms | tok/sec: 879,145 | bf16_mfu: 54.00 | total time: 10.00m | eta: 102.1m step 00507/05568 (9.11%) | loss: 3.000700 | lrm: 1.00 | dt: 1199.82ms | tok/sec: 873,945 | bf16_mfu: 53.68 | total time: 10.02m | eta: 102.1m step 00508/05568 (9.12%) | loss: 2.998691 | lrm: 1.00 | dt: 1198.80ms | tok/sec: 874,686 | bf16_mfu: 53.73 | total time: 10.04m | eta: 102.0m step 00509/05568 (9.14%) | loss: 2.993619 | lrm: 1.00 | dt: 1197.41ms | tok/sec: 875,700 | bf16_mfu: 53.79 | total time: 10.06m | eta: 102.0m step 00510/05568 (9.16%) | loss: 3.000147 | lrm: 1.00 | dt: 1203.27ms | tok/sec: 871,441 | bf16_mfu: 53.53 | total time: 10.08m | eta: 102.0m step 00511/05568 (9.18%) | loss: 3.004602 | lrm: 1.00 | dt: 1199.25ms | tok/sec: 874,358 | bf16_mfu: 53.71 | total time: 10.10m | eta: 102.0m step 00512/05568 (9.20%) | loss: 3.012331 | lrm: 1.00 | dt: 1195.47ms | tok/sec: 877,120 | bf16_mfu: 53.88 | total time: 10.12m | eta: 101.9m step 00513/05568 (9.21%) | loss: 3.014264 | lrm: 1.00 | dt: 1204.50ms | tok/sec: 870,548 | bf16_mfu: 53.48 | total time: 10.14m | eta: 101.9m step 00514/05568 (9.23%) | loss: 3.015397 | lrm: 1.00 | dt: 1199.66ms | tok/sec: 874,064 | bf16_mfu: 53.69 | total time: 10.16m | eta: 101.9m step 00515/05568 (9.25%) | loss: 3.005792 | lrm: 1.00 | dt: 1200.76ms | tok/sec: 873,257 | bf16_mfu: 53.64 | total time: 10.18m | eta: 101.9m step 00516/05568 (9.27%) | loss: 3.002888 | lrm: 1.00 | dt: 1198.98ms | tok/sec: 874,555 | bf16_mfu: 53.72 | total time: 10.20m | eta: 101.9m step 00517/05568 (9.29%) | loss: 2.988147 | lrm: 1.00 | dt: 1195.08ms | tok/sec: 877,413 | bf16_mfu: 53.90 | total time: 10.22m | eta: 101.8m step 00518/05568 (9.30%) | loss: 2.978779 | lrm: 1.00 | dt: 1204.15ms | tok/sec: 870,802 | bf16_mfu: 53.49 | total time: 10.24m | eta: 101.8m step 00519/05568 (9.32%) | loss: 2.973171 | lrm: 1.00 | dt: 1195.53ms | tok/sec: 877,081 | bf16_mfu: 53.88 | total time: 10.26m | eta: 101.8m step 00520/05568 (9.34%) | loss: 2.970520 | lrm: 1.00 | dt: 1200.41ms | tok/sec: 873,513 | bf16_mfu: 53.66 | total time: 10.28m | eta: 101.8m step 00521/05568 (9.36%) | loss: 2.967344 | lrm: 1.00 | dt: 1195.22ms | tok/sec: 877,307 | bf16_mfu: 53.89 | total time: 10.30m | eta: 101.8m step 00522/05568 (9.38%) | loss: 2.968373 | lrm: 1.00 | dt: 1195.98ms | tok/sec: 876,749 | bf16_mfu: 53.86 | total time: 10.32m | eta: 101.7m step 00523/05568 (9.39%) | loss: 2.959791 | lrm: 1.00 | dt: 1194.00ms | tok/sec: 878,200 | bf16_mfu: 53.95 | total time: 10.34m | eta: 101.7m step 00524/05568 (9.41%) | loss: 2.955204 | lrm: 1.00 | dt: 1199.86ms | tok/sec: 873,918 | bf16_mfu: 53.68 | total time: 10.36m | eta: 101.7m step 00525/05568 (9.43%) | loss: 2.962749 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,637 | bf16_mfu: 53.91 | total time: 10.38m | eta: 101.7m step 00526/05568 (9.45%) | loss: 2.966729 | lrm: 1.00 | dt: 1189.50ms | tok/sec: 881,529 | bf16_mfu: 54.15 | total time: 10.40m | eta: 101.6m step 00527/05568 (9.46%) | loss: 2.965206 | lrm: 1.00 | dt: 1205.08ms | tok/sec: 870,129 | bf16_mfu: 53.45 | total time: 10.42m | eta: 101.6m step 00528/05568 (9.48%) | loss: 2.962575 | lrm: 1.00 | dt: 1196.66ms | tok/sec: 876,251 | bf16_mfu: 53.83 | total time: 10.44m | eta: 101.6m step 00529/05568 (9.50%) | loss: 2.957466 | lrm: 1.00 | dt: 1195.67ms | tok/sec: 876,980 | bf16_mfu: 53.87 | total time: 10.46m | eta: 101.6m step 00530/05568 (9.52%) | loss: 2.965908 | lrm: 1.00 | dt: 1194.69ms | tok/sec: 877,694 | bf16_mfu: 53.91 | total time: 10.48m | eta: 101.6m step 00531/05568 (9.54%) | loss: 2.964021 | lrm: 1.00 | dt: 1196.34ms | tok/sec: 876,488 | bf16_mfu: 53.84 | total time: 10.50m | eta: 101.5m step 00532/05568 (9.55%) | loss: 2.962372 | lrm: 1.00 | dt: 1197.36ms | tok/sec: 875,741 | bf16_mfu: 53.79 | total time: 10.52m | eta: 101.5m step 00533/05568 (9.57%) | loss: 2.973026 | lrm: 1.00 | dt: 1196.87ms | tok/sec: 876,100 | bf16_mfu: 53.82 | total time: 10.54m | eta: 101.5m step 00534/05568 (9.59%) | loss: 2.969515 | lrm: 1.00 | dt: 1198.10ms | tok/sec: 875,201 | bf16_mfu: 53.76 | total time: 10.56m | eta: 101.5m step 00535/05568 (9.61%) | loss: 2.953894 | lrm: 1.00 | dt: 1191.99ms | tok/sec: 879,685 | bf16_mfu: 54.04 | total time: 10.58m | eta: 101.4m step 00536/05568 (9.63%) | loss: 2.953212 | lrm: 1.00 | dt: 1193.75ms | tok/sec: 878,388 | bf16_mfu: 53.96 | total time: 10.60m | eta: 101.4m step 00537/05568 (9.64%) | loss: 2.957726 | lrm: 1.00 | dt: 1197.29ms | tok/sec: 875,790 | bf16_mfu: 53.80 | total time: 10.62m | eta: 101.4m step 00538/05568 (9.66%) | loss: 2.943805 | lrm: 1.00 | dt: 1198.57ms | tok/sec: 874,852 | bf16_mfu: 53.74 | total time: 10.64m | eta: 101.4m step 00539/05568 (9.68%) | loss: 2.941704 | lrm: 1.00 | dt: 1195.19ms | tok/sec: 877,330 | bf16_mfu: 53.89 | total time: 10.66m | eta: 101.4m step 00540/05568 (9.70%) | loss: 2.921479 | lrm: 1.00 | dt: 1196.94ms | tok/sec: 876,044 | bf16_mfu: 53.81 | total time: 10.68m | eta: 101.3m step 00541/05568 (9.72%) | loss: 2.927295 | lrm: 1.00 | dt: 1198.65ms | tok/sec: 874,799 | bf16_mfu: 53.74 | total time: 10.70m | eta: 101.3m step 00542/05568 (9.73%) | loss: 2.946763 | lrm: 1.00 | dt: 1196.32ms | tok/sec: 876,504 | bf16_mfu: 53.84 | total time: 10.72m | eta: 101.3m step 00543/05568 (9.75%) | loss: 2.937368 | lrm: 1.00 | dt: 1197.88ms | tok/sec: 875,356 | bf16_mfu: 53.77 | total time: 10.74m | eta: 101.3m step 00544/05568 (9.77%) | loss: 2.939183 | lrm: 1.00 | dt: 1199.30ms | tok/sec: 874,320 | bf16_mfu: 53.71 | total time: 10.76m | eta: 101.2m step 00545/05568 (9.79%) | loss: 2.937466 | lrm: 1.00 | dt: 1199.24ms | tok/sec: 874,366 | bf16_mfu: 53.71 | total time: 10.78m | eta: 101.2m step 00546/05568 (9.81%) | loss: 2.930958 | lrm: 1.00 | dt: 1197.48ms | tok/sec: 875,654 | bf16_mfu: 53.79 | total time: 10.80m | eta: 101.2m step 00547/05568 (9.82%) | loss: 2.930409 | lrm: 1.00 | dt: 1200.12ms | tok/sec: 873,723 | bf16_mfu: 53.67 | total time: 10.82m | eta: 101.2m step 00548/05568 (9.84%) | loss: 2.923070 | lrm: 1.00 | dt: 1199.71ms | tok/sec: 874,025 | bf16_mfu: 53.69 | total time: 10.84m | eta: 101.2m step 00549/05568 (9.86%) | loss: 2.920006 | lrm: 1.00 | dt: 1197.00ms | tok/sec: 876,005 | bf16_mfu: 53.81 | total time: 10.86m | eta: 101.1m step 00550/05568 (9.88%) | loss: 2.918409 | lrm: 1.00 | dt: 1195.90ms | tok/sec: 876,805 | bf16_mfu: 53.86 | total time: 10.88m | eta: 101.1m step 00551/05568 (9.90%) | loss: 2.930751 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,263 | bf16_mfu: 53.95 | total time: 10.90m | eta: 101.1m step 00552/05568 (9.91%) | loss: 2.929497 | lrm: 1.00 | dt: 1196.09ms | tok/sec: 876,668 | bf16_mfu: 53.85 | total time: 10.92m | eta: 101.1m step 00553/05568 (9.93%) | loss: 2.942040 | lrm: 1.00 | dt: 1193.51ms | tok/sec: 878,561 | bf16_mfu: 53.97 | total time: 10.94m | eta: 101.0m step 00554/05568 (9.95%) | loss: 2.948819 | lrm: 1.00 | dt: 1202.58ms | tok/sec: 871,941 | bf16_mfu: 53.56 | total time: 10.96m | eta: 101.0m step 00555/05568 (9.97%) | loss: 2.957710 | lrm: 1.00 | dt: 1194.93ms | tok/sec: 877,519 | bf16_mfu: 53.90 | total time: 10.98m | eta: 101.0m step 00556/05568 (9.99%) | loss: 2.962084 | lrm: 1.00 | dt: 1192.20ms | tok/sec: 879,528 | bf16_mfu: 54.03 | total time: 11.00m | eta: 101.0m step 00557/05568 (10.00%) | loss: 2.970806 | lrm: 1.00 | dt: 1193.81ms | tok/sec: 878,346 | bf16_mfu: 53.95 | total time: 11.02m | eta: 101.0m step 00558/05568 (10.02%) | loss: 2.955618 | lrm: 1.00 | dt: 1199.23ms | tok/sec: 874,373 | bf16_mfu: 53.71 | total time: 11.04m | eta: 100.9m step 00559/05568 (10.04%) | loss: 2.944165 | lrm: 1.00 | dt: 1195.63ms | tok/sec: 877,005 | bf16_mfu: 53.87 | total time: 11.06m | eta: 100.9m step 00560/05568 (10.06%) | loss: 2.957334 | lrm: 1.00 | dt: 1197.12ms | tok/sec: 875,918 | bf16_mfu: 53.81 | total time: 11.08m | eta: 100.9m step 00561/05568 (10.08%) | loss: 2.945573 | lrm: 1.00 | dt: 1195.40ms | tok/sec: 877,177 | bf16_mfu: 53.88 | total time: 11.10m | eta: 100.9m step 00562/05568 (10.09%) | loss: 2.948755 | lrm: 1.00 | dt: 1191.64ms | tok/sec: 879,945 | bf16_mfu: 54.05 | total time: 11.12m | eta: 100.8m step 00563/05568 (10.11%) | loss: 2.949893 | lrm: 1.00 | dt: 1197.84ms | tok/sec: 875,389 | bf16_mfu: 53.77 | total time: 11.14m | eta: 100.8m step 00564/05568 (10.13%) | loss: 2.940571 | lrm: 1.00 | dt: 1196.73ms | tok/sec: 876,200 | bf16_mfu: 53.82 | total time: 11.16m | eta: 100.8m step 00565/05568 (10.15%) | loss: 2.946856 | lrm: 1.00 | dt: 1196.62ms | tok/sec: 876,284 | bf16_mfu: 53.83 | total time: 11.18m | eta: 100.8m step 00566/05568 (10.17%) | loss: 2.944247 | lrm: 1.00 | dt: 1194.61ms | tok/sec: 877,756 | bf16_mfu: 53.92 | total time: 11.20m | eta: 100.8m step 00567/05568 (10.18%) | loss: 2.937374 | lrm: 1.00 | dt: 1193.77ms | tok/sec: 878,373 | bf16_mfu: 53.96 | total time: 11.22m | eta: 100.7m step 00568/05568 (10.20%) | loss: 2.940442 | lrm: 1.00 | dt: 1192.91ms | tok/sec: 879,007 | bf16_mfu: 54.00 | total time: 11.24m | eta: 100.7m step 00569/05568 (10.22%) | loss: 2.947135 | lrm: 1.00 | dt: 1196.69ms | tok/sec: 876,231 | bf16_mfu: 53.82 | total time: 11.26m | eta: 100.7m step 00570/05568 (10.24%) | loss: 2.944988 | lrm: 1.00 | dt: 1196.75ms | tok/sec: 876,188 | bf16_mfu: 53.82 | total time: 11.28m | eta: 100.7m step 00571/05568 (10.26%) | loss: 2.943608 | lrm: 1.00 | dt: 1193.40ms | tok/sec: 878,643 | bf16_mfu: 53.97 | total time: 11.30m | eta: 100.6m step 00572/05568 (10.27%) | loss: 2.922054 | lrm: 1.00 | dt: 1197.59ms | tok/sec: 875,570 | bf16_mfu: 53.78 | total time: 11.32m | eta: 100.6m step 00573/05568 (10.29%) | loss: 2.912733 | lrm: 1.00 | dt: 1192.24ms | tok/sec: 879,502 | bf16_mfu: 54.03 | total time: 11.34m | eta: 100.6m step 00574/05568 (10.31%) | loss: 2.909240 | lrm: 1.00 | dt: 1194.08ms | tok/sec: 878,144 | bf16_mfu: 53.94 | total time: 11.36m | eta: 100.6m step 00575/05568 (10.33%) | loss: 2.909636 | lrm: 1.00 | dt: 1200.48ms | tok/sec: 873,461 | bf16_mfu: 53.65 | total time: 11.38m | eta: 100.6m step 00576/05568 (10.34%) | loss: 2.929341 | lrm: 1.00 | dt: 1202.14ms | tok/sec: 872,254 | bf16_mfu: 53.58 | total time: 11.40m | eta: 100.5m step 00577/05568 (10.36%) | loss: 2.936234 | lrm: 1.00 | dt: 1193.54ms | tok/sec: 878,543 | bf16_mfu: 53.97 | total time: 11.42m | eta: 100.5m step 00578/05568 (10.38%) | loss: 2.929870 | lrm: 1.00 | dt: 1198.08ms | tok/sec: 875,215 | bf16_mfu: 53.76 | total time: 11.44m | eta: 100.5m step 00579/05568 (10.40%) | loss: 2.926554 | lrm: 1.00 | dt: 1201.33ms | tok/sec: 872,843 | bf16_mfu: 53.62 | total time: 11.46m | eta: 100.5m step 00580/05568 (10.42%) | loss: 2.919644 | lrm: 1.00 | dt: 1189.04ms | tok/sec: 881,864 | bf16_mfu: 54.17 | total time: 11.48m | eta: 100.4m step 00581/05568 (10.43%) | loss: 2.909696 | lrm: 1.00 | dt: 1200.02ms | tok/sec: 873,796 | bf16_mfu: 53.68 | total time: 11.50m | eta: 100.4m step 00582/05568 (10.45%) | loss: 2.904975 | lrm: 1.00 | dt: 1192.29ms | tok/sec: 879,464 | bf16_mfu: 54.02 | total time: 11.52m | eta: 100.4m step 00583/05568 (10.47%) | loss: 2.919984 | lrm: 1.00 | dt: 1200.77ms | tok/sec: 873,255 | bf16_mfu: 53.64 | total time: 11.54m | eta: 100.4m step 00584/05568 (10.49%) | loss: 2.925510 | lrm: 1.00 | dt: 1197.37ms | tok/sec: 875,729 | bf16_mfu: 53.79 | total time: 11.56m | eta: 100.4m step 00585/05568 (10.51%) | loss: 2.923050 | lrm: 1.00 | dt: 1194.92ms | tok/sec: 877,529 | bf16_mfu: 53.90 | total time: 11.58m | eta: 100.3m step 00586/05568 (10.52%) | loss: 2.924843 | lrm: 1.00 | dt: 1202.10ms | tok/sec: 872,287 | bf16_mfu: 53.58 | total time: 11.60m | eta: 100.3m step 00587/05568 (10.54%) | loss: 2.929189 | lrm: 1.00 | dt: 1201.84ms | tok/sec: 872,479 | bf16_mfu: 53.59 | total time: 11.62m | eta: 100.3m step 00588/05568 (10.56%) | loss: 2.935670 | lrm: 1.00 | dt: 1193.22ms | tok/sec: 878,782 | bf16_mfu: 53.98 | total time: 11.64m | eta: 100.3m step 00589/05568 (10.58%) | loss: 2.948012 | lrm: 1.00 | dt: 1201.94ms | tok/sec: 872,405 | bf16_mfu: 53.59 | total time: 11.66m | eta: 100.3m step 00590/05568 (10.60%) | loss: 2.955193 | lrm: 1.00 | dt: 1197.95ms | tok/sec: 875,305 | bf16_mfu: 53.77 | total time: 11.68m | eta: 100.2m step 00591/05568 (10.61%) | loss: 2.948546 | lrm: 1.00 | dt: 1191.68ms | tok/sec: 879,916 | bf16_mfu: 54.05 | total time: 11.70m | eta: 100.2m step 00592/05568 (10.63%) | loss: 2.945737 | lrm: 1.00 | dt: 1194.08ms | tok/sec: 878,145 | bf16_mfu: 53.94 | total time: 11.72m | eta: 100.2m step 00593/05568 (10.65%) | loss: 2.935314 | lrm: 1.00 | dt: 1196.60ms | tok/sec: 876,296 | bf16_mfu: 53.83 | total time: 11.74m | eta: 100.2m step 00594/05568 (10.67%) | loss: 2.940034 | lrm: 1.00 | dt: 1200.12ms | tok/sec: 873,723 | bf16_mfu: 53.67 | total time: 11.76m | eta: 100.1m step 00595/05568 (10.69%) | loss: 2.935977 | lrm: 1.00 | dt: 1196.68ms | tok/sec: 876,239 | bf16_mfu: 53.83 | total time: 11.78m | eta: 100.1m step 00596/05568 (10.70%) | loss: 2.943329 | lrm: 1.00 | dt: 1194.71ms | tok/sec: 877,681 | bf16_mfu: 53.91 | total time: 11.80m | eta: 100.1m step 00597/05568 (10.72%) | loss: 2.933517 | lrm: 1.00 | dt: 1193.98ms | tok/sec: 878,221 | bf16_mfu: 53.95 | total time: 11.82m | eta: 100.1m step 00598/05568 (10.74%) | loss: 2.931315 | lrm: 1.00 | dt: 1199.27ms | tok/sec: 874,347 | bf16_mfu: 53.71 | total time: 11.84m | eta: 100.1m step 00599/05568 (10.76%) | loss: 2.933736 | lrm: 1.00 | dt: 1189.98ms | tok/sec: 881,173 | bf16_mfu: 54.13 | total time: 11.86m | eta: 100.0m step 00600/05568 (10.78%) | loss: 2.922162 | lrm: 1.00 | dt: 1199.07ms | tok/sec: 874,492 | bf16_mfu: 53.72 | total time: 11.88m | eta: 100.0m step 00601/05568 (10.79%) | loss: 2.921615 | lrm: 1.00 | dt: 1194.37ms | tok/sec: 877,931 | bf16_mfu: 53.93 | total time: 11.90m | eta: 100.0m step 00602/05568 (10.81%) | loss: 2.913513 | lrm: 1.00 | dt: 1196.29ms | tok/sec: 876,523 | bf16_mfu: 53.84 | total time: 11.92m | eta: 100.0m step 00603/05568 (10.83%) | loss: 2.921916 | lrm: 1.00 | dt: 1197.70ms | tok/sec: 875,488 | bf16_mfu: 53.78 | total time: 11.94m | eta: 99.9m step 00604/05568 (10.85%) | loss: 2.919176 | lrm: 1.00 | dt: 1204.37ms | tok/sec: 870,645 | bf16_mfu: 53.48 | total time: 11.96m | eta: 99.9m step 00605/05568 (10.87%) | loss: 2.917187 | lrm: 1.00 | dt: 1193.85ms | tok/sec: 878,317 | bf16_mfu: 53.95 | total time: 11.98m | eta: 99.9m step 00606/05568 (10.88%) | loss: 2.918680 | lrm: 1.00 | dt: 1194.49ms | tok/sec: 877,844 | bf16_mfu: 53.92 | total time: 12.00m | eta: 99.9m step 00607/05568 (10.90%) | loss: 2.918466 | lrm: 1.00 | dt: 1193.63ms | tok/sec: 878,479 | bf16_mfu: 53.96 | total time: 12.02m | eta: 99.9m step 00608/05568 (10.92%) | loss: 2.914252 | lrm: 1.00 | dt: 1193.31ms | tok/sec: 878,709 | bf16_mfu: 53.98 | total time: 12.04m | eta: 99.8m step 00609/05568 (10.94%) | loss: 2.903479 | lrm: 1.00 | dt: 1197.02ms | tok/sec: 875,990 | bf16_mfu: 53.81 | total time: 12.06m | eta: 99.8m step 00610/05568 (10.96%) | loss: 2.906867 | lrm: 1.00 | dt: 1191.93ms | tok/sec: 879,732 | bf16_mfu: 54.04 | total time: 12.08m | eta: 99.8m step 00611/05568 (10.97%) | loss: 2.916792 | lrm: 1.00 | dt: 1196.38ms | tok/sec: 876,458 | bf16_mfu: 53.84 | total time: 12.10m | eta: 99.8m step 00612/05568 (10.99%) | loss: 2.907302 | lrm: 1.00 | dt: 1203.91ms | tok/sec: 870,977 | bf16_mfu: 53.50 | total time: 12.12m | eta: 99.7m step 00613/05568 (11.01%) | loss: 2.901651 | lrm: 1.00 | dt: 1198.99ms | tok/sec: 874,546 | bf16_mfu: 53.72 | total time: 12.14m | eta: 99.7m step 00614/05568 (11.03%) | loss: 2.889304 | lrm: 1.00 | dt: 1194.04ms | tok/sec: 878,172 | bf16_mfu: 53.94 | total time: 12.16m | eta: 99.7m step 00615/05568 (11.05%) | loss: 2.901253 | lrm: 1.00 | dt: 1193.82ms | tok/sec: 878,337 | bf16_mfu: 53.95 | total time: 12.18m | eta: 99.7m step 00616/05568 (11.06%) | loss: 2.897667 | lrm: 1.00 | dt: 1193.04ms | tok/sec: 878,910 | bf16_mfu: 53.99 | total time: 12.20m | eta: 99.7m step 00617/05568 (11.08%) | loss: 2.902754 | lrm: 1.00 | dt: 1191.53ms | tok/sec: 880,024 | bf16_mfu: 54.06 | total time: 12.22m | eta: 99.6m step 00618/05568 (11.10%) | loss: 2.911297 | lrm: 1.00 | dt: 1195.23ms | tok/sec: 877,302 | bf16_mfu: 53.89 | total time: 12.24m | eta: 99.6m step 00619/05568 (11.12%) | loss: 2.910280 | lrm: 1.00 | dt: 1197.93ms | tok/sec: 875,323 | bf16_mfu: 53.77 | total time: 12.26m | eta: 99.6m step 00620/05568 (11.14%) | loss: 2.906378 | lrm: 1.00 | dt: 1201.40ms | tok/sec: 872,798 | bf16_mfu: 53.61 | total time: 12.28m | eta: 99.6m step 00621/05568 (11.15%) | loss: 2.905681 | lrm: 1.00 | dt: 1199.53ms | tok/sec: 874,154 | bf16_mfu: 53.70 | total time: 12.30m | eta: 99.6m step 00622/05568 (11.17%) | loss: 2.901166 | lrm: 1.00 | dt: 1199.09ms | tok/sec: 874,478 | bf16_mfu: 53.72 | total time: 12.32m | eta: 99.5m step 00623/05568 (11.19%) | loss: 2.904332 | lrm: 1.00 | dt: 1191.76ms | tok/sec: 879,852 | bf16_mfu: 54.05 | total time: 12.34m | eta: 99.5m step 00624/05568 (11.21%) | loss: 2.905055 | lrm: 1.00 | dt: 1197.52ms | tok/sec: 875,620 | bf16_mfu: 53.79 | total time: 12.36m | eta: 99.5m step 00625/05568 (11.22%) | loss: 2.907147 | lrm: 1.00 | dt: 1195.91ms | tok/sec: 876,798 | bf16_mfu: 53.86 | total time: 12.38m | eta: 99.5m step 00626/05568 (11.24%) | loss: 2.910348 | lrm: 1.00 | dt: 1189.93ms | tok/sec: 881,209 | bf16_mfu: 54.13 | total time: 12.40m | eta: 99.4m step 00627/05568 (11.26%) | loss: 2.897776 | lrm: 1.00 | dt: 1197.99ms | tok/sec: 875,278 | bf16_mfu: 53.77 | total time: 12.42m | eta: 99.4m step 00628/05568 (11.28%) | loss: 2.909938 | lrm: 1.00 | dt: 1197.62ms | tok/sec: 875,549 | bf16_mfu: 53.78 | total time: 12.44m | eta: 99.4m step 00629/05568 (11.30%) | loss: 2.911034 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,396 | bf16_mfu: 54.02 | total time: 12.46m | eta: 99.4m step 00630/05568 (11.31%) | loss: 2.907482 | lrm: 1.00 | dt: 1195.69ms | tok/sec: 876,962 | bf16_mfu: 53.87 | total time: 12.48m | eta: 99.4m step 00631/05568 (11.33%) | loss: 2.904097 | lrm: 1.00 | dt: 1190.63ms | tok/sec: 880,691 | bf16_mfu: 54.10 | total time: 12.50m | eta: 99.3m step 00632/05568 (11.35%) | loss: 2.906087 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,304 | bf16_mfu: 53.95 | total time: 12.52m | eta: 99.3m step 00633/05568 (11.37%) | loss: 2.907143 | lrm: 1.00 | dt: 1197.34ms | tok/sec: 875,755 | bf16_mfu: 53.80 | total time: 12.54m | eta: 99.3m step 00634/05568 (11.39%) | loss: 2.913204 | lrm: 1.00 | dt: 1201.98ms | tok/sec: 872,373 | bf16_mfu: 53.59 | total time: 12.56m | eta: 99.3m step 00635/05568 (11.40%) | loss: 2.915286 | lrm: 1.00 | dt: 1194.95ms | tok/sec: 877,506 | bf16_mfu: 53.90 | total time: 12.58m | eta: 99.2m step 00636/05568 (11.42%) | loss: 2.910066 | lrm: 1.00 | dt: 1195.93ms | tok/sec: 876,787 | bf16_mfu: 53.86 | total time: 12.60m | eta: 99.2m step 00637/05568 (11.44%) | loss: 2.911860 | lrm: 1.00 | dt: 1192.49ms | tok/sec: 879,319 | bf16_mfu: 54.01 | total time: 12.62m | eta: 99.2m step 00638/05568 (11.46%) | loss: 2.922294 | lrm: 1.00 | dt: 1193.80ms | tok/sec: 878,353 | bf16_mfu: 53.96 | total time: 12.64m | eta: 99.2m step 00639/05568 (11.48%) | loss: 2.914880 | lrm: 1.00 | dt: 1196.49ms | tok/sec: 876,375 | bf16_mfu: 53.83 | total time: 12.66m | eta: 99.2m step 00640/05568 (11.49%) | loss: 2.913295 | lrm: 1.00 | dt: 1196.31ms | tok/sec: 876,506 | bf16_mfu: 53.84 | total time: 12.68m | eta: 99.1m step 00641/05568 (11.51%) | loss: 2.903540 | lrm: 1.00 | dt: 1195.76ms | tok/sec: 876,910 | bf16_mfu: 53.87 | total time: 12.70m | eta: 99.1m step 00642/05568 (11.53%) | loss: 2.897842 | lrm: 1.00 | dt: 1195.32ms | tok/sec: 877,234 | bf16_mfu: 53.89 | total time: 12.72m | eta: 99.1m step 00643/05568 (11.55%) | loss: 2.894673 | lrm: 1.00 | dt: 1199.93ms | tok/sec: 873,862 | bf16_mfu: 53.68 | total time: 12.74m | eta: 99.1m step 00644/05568 (11.57%) | loss: 2.890442 | lrm: 1.00 | dt: 1196.41ms | tok/sec: 876,437 | bf16_mfu: 53.84 | total time: 12.76m | eta: 99.1m step 00645/05568 (11.58%) | loss: 2.885412 | lrm: 1.00 | dt: 1197.53ms | tok/sec: 875,618 | bf16_mfu: 53.79 | total time: 12.78m | eta: 99.0m step 00646/05568 (11.60%) | loss: 2.886350 | lrm: 1.00 | dt: 1194.80ms | tok/sec: 877,614 | bf16_mfu: 53.91 | total time: 12.80m | eta: 99.0m step 00647/05568 (11.62%) | loss: 2.881861 | lrm: 1.00 | dt: 1197.40ms | tok/sec: 875,710 | bf16_mfu: 53.79 | total time: 12.82m | eta: 99.0m step 00648/05568 (11.64%) | loss: 2.877095 | lrm: 1.00 | dt: 1191.60ms | tok/sec: 879,976 | bf16_mfu: 54.05 | total time: 12.84m | eta: 99.0m step 00649/05568 (11.66%) | loss: 2.881522 | lrm: 1.00 | dt: 1198.88ms | tok/sec: 874,626 | bf16_mfu: 53.73 | total time: 12.86m | eta: 98.9m step 00650/05568 (11.67%) | loss: 2.880440 | lrm: 1.00 | dt: 1201.73ms | tok/sec: 872,552 | bf16_mfu: 53.60 | total time: 12.88m | eta: 98.9m step 00651/05568 (11.69%) | loss: 2.889746 | lrm: 1.00 | dt: 1200.56ms | tok/sec: 873,405 | bf16_mfu: 53.65 | total time: 12.90m | eta: 98.9m step 00652/05568 (11.71%) | loss: 2.883894 | lrm: 1.00 | dt: 1193.38ms | tok/sec: 878,663 | bf16_mfu: 53.97 | total time: 12.92m | eta: 98.9m step 00653/05568 (11.73%) | loss: 2.882527 | lrm: 1.00 | dt: 1191.94ms | tok/sec: 879,720 | bf16_mfu: 54.04 | total time: 12.94m | eta: 98.9m step 00654/05568 (11.75%) | loss: 2.896971 | lrm: 1.00 | dt: 1196.79ms | tok/sec: 876,156 | bf16_mfu: 53.82 | total time: 12.96m | eta: 98.8m step 00655/05568 (11.76%) | loss: 2.889247 | lrm: 1.00 | dt: 1200.53ms | tok/sec: 873,427 | bf16_mfu: 53.65 | total time: 12.98m | eta: 98.8m step 00656/05568 (11.78%) | loss: 2.896052 | lrm: 1.00 | dt: 1199.41ms | tok/sec: 874,240 | bf16_mfu: 53.70 | total time: 13.00m | eta: 98.8m step 00657/05568 (11.80%) | loss: 2.895777 | lrm: 1.00 | dt: 1193.81ms | tok/sec: 878,347 | bf16_mfu: 53.95 | total time: 13.02m | eta: 98.8m step 00658/05568 (11.82%) | loss: 2.891035 | lrm: 1.00 | dt: 1193.07ms | tok/sec: 878,887 | bf16_mfu: 53.99 | total time: 13.04m | eta: 98.8m step 00659/05568 (11.84%) | loss: 2.890908 | lrm: 1.00 | dt: 1199.04ms | tok/sec: 874,514 | bf16_mfu: 53.72 | total time: 13.06m | eta: 98.7m step 00660/05568 (11.85%) | loss: 2.888707 | lrm: 1.00 | dt: 1194.36ms | tok/sec: 877,940 | bf16_mfu: 53.93 | total time: 13.08m | eta: 98.7m step 00661/05568 (11.87%) | loss: 2.882636 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,266 | bf16_mfu: 53.95 | total time: 13.10m | eta: 98.7m step 00662/05568 (11.89%) | loss: 2.879635 | lrm: 1.00 | dt: 1194.39ms | tok/sec: 877,915 | bf16_mfu: 53.93 | total time: 13.12m | eta: 98.7m step 00663/05568 (11.91%) | loss: 2.869769 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,310 | bf16_mfu: 53.95 | total time: 13.14m | eta: 98.6m step 00664/05568 (11.93%) | loss: 2.866411 | lrm: 1.00 | dt: 1192.48ms | tok/sec: 879,324 | bf16_mfu: 54.01 | total time: 13.16m | eta: 98.6m step 00665/05568 (11.94%) | loss: 2.858255 | lrm: 1.00 | dt: 1197.13ms | tok/sec: 875,904 | bf16_mfu: 53.80 | total time: 13.18m | eta: 98.6m step 00666/05568 (11.96%) | loss: 2.849788 | lrm: 1.00 | dt: 1196.65ms | tok/sec: 876,256 | bf16_mfu: 53.83 | total time: 13.20m | eta: 98.6m step 00667/05568 (11.98%) | loss: 2.853488 | lrm: 1.00 | dt: 1195.33ms | tok/sec: 877,229 | bf16_mfu: 53.89 | total time: 13.22m | eta: 98.6m step 00668/05568 (12.00%) | loss: 2.868557 | lrm: 1.00 | dt: 1200.29ms | tok/sec: 873,604 | bf16_mfu: 53.66 | total time: 13.24m | eta: 98.5m step 00669/05568 (12.02%) | loss: 2.870184 | lrm: 1.00 | dt: 1193.19ms | tok/sec: 878,797 | bf16_mfu: 53.98 | total time: 13.26m | eta: 98.5m step 00670/05568 (12.03%) | loss: 2.878927 | lrm: 1.00 | dt: 1194.62ms | tok/sec: 877,746 | bf16_mfu: 53.92 | total time: 13.27m | eta: 98.5m step 00671/05568 (12.05%) | loss: 2.872720 | lrm: 1.00 | dt: 1193.91ms | tok/sec: 878,266 | bf16_mfu: 53.95 | total time: 13.29m | eta: 98.5m step 00672/05568 (12.07%) | loss: 2.878440 | lrm: 1.00 | dt: 1193.82ms | tok/sec: 878,334 | bf16_mfu: 53.95 | total time: 13.31m | eta: 98.5m step 00673/05568 (12.09%) | loss: 2.877662 | lrm: 1.00 | dt: 1193.50ms | tok/sec: 878,570 | bf16_mfu: 53.97 | total time: 13.33m | eta: 98.4m step 00674/05568 (12.10%) | loss: 2.873920 | lrm: 1.00 | dt: 1193.78ms | tok/sec: 878,365 | bf16_mfu: 53.96 | total time: 13.35m | eta: 98.4m step 00675/05568 (12.12%) | loss: 2.868292 | lrm: 1.00 | dt: 1198.10ms | tok/sec: 875,197 | bf16_mfu: 53.76 | total time: 13.37m | eta: 98.4m step 00676/05568 (12.14%) | loss: 2.878717 | lrm: 1.00 | dt: 1192.27ms | tok/sec: 879,475 | bf16_mfu: 54.02 | total time: 13.39m | eta: 98.4m step 00677/05568 (12.16%) | loss: 2.878687 | lrm: 1.00 | dt: 1197.49ms | tok/sec: 875,644 | bf16_mfu: 53.79 | total time: 13.41m | eta: 98.3m step 00678/05568 (12.18%) | loss: 2.872178 | lrm: 1.00 | dt: 1194.59ms | tok/sec: 877,769 | bf16_mfu: 53.92 | total time: 13.43m | eta: 98.3m step 00679/05568 (12.19%) | loss: 2.873884 | lrm: 1.00 | dt: 1192.42ms | tok/sec: 879,370 | bf16_mfu: 54.02 | total time: 13.45m | eta: 98.3m step 00680/05568 (12.21%) | loss: 2.872768 | lrm: 1.00 | dt: 1195.06ms | tok/sec: 877,425 | bf16_mfu: 53.90 | total time: 13.47m | eta: 98.3m step 00681/05568 (12.23%) | loss: 2.866219 | lrm: 1.00 | dt: 1196.81ms | tok/sec: 876,140 | bf16_mfu: 53.82 | total time: 13.49m | eta: 98.3m step 00682/05568 (12.25%) | loss: 2.849476 | lrm: 1.00 | dt: 1192.07ms | tok/sec: 879,627 | bf16_mfu: 54.03 | total time: 13.51m | eta: 98.2m step 00683/05568 (12.27%) | loss: 2.863749 | lrm: 1.00 | dt: 1191.26ms | tok/sec: 880,222 | bf16_mfu: 54.07 | total time: 13.53m | eta: 98.2m step 00684/05568 (12.28%) | loss: 2.867078 | lrm: 1.00 | dt: 1197.39ms | tok/sec: 875,716 | bf16_mfu: 53.79 | total time: 13.55m | eta: 98.2m step 00685/05568 (12.30%) | loss: 2.874663 | lrm: 1.00 | dt: 1191.57ms | tok/sec: 879,998 | bf16_mfu: 54.06 | total time: 13.57m | eta: 98.2m step 00686/05568 (12.32%) | loss: 2.875846 | lrm: 1.00 | dt: 1199.53ms | tok/sec: 874,158 | bf16_mfu: 53.70 | total time: 13.59m | eta: 98.2m step 00687/05568 (12.34%) | loss: 2.867481 | lrm: 1.00 | dt: 1195.85ms | tok/sec: 876,845 | bf16_mfu: 53.86 | total time: 13.61m | eta: 98.1m step 00688/05568 (12.36%) | loss: 2.859199 | lrm: 1.00 | dt: 1198.99ms | tok/sec: 874,548 | bf16_mfu: 53.72 | total time: 13.63m | eta: 98.1m step 00689/05568 (12.37%) | loss: 2.848739 | lrm: 1.00 | dt: 1196.51ms | tok/sec: 876,361 | bf16_mfu: 53.83 | total time: 13.65m | eta: 98.1m step 00690/05568 (12.39%) | loss: 2.846167 | lrm: 1.00 | dt: 1193.58ms | tok/sec: 878,516 | bf16_mfu: 53.97 | total time: 13.67m | eta: 98.1m step 00691/05568 (12.41%) | loss: 2.849650 | lrm: 1.00 | dt: 1191.23ms | tok/sec: 880,249 | bf16_mfu: 54.07 | total time: 13.69m | eta: 98.0m step 00692/05568 (12.43%) | loss: 2.862644 | lrm: 1.00 | dt: 1195.27ms | tok/sec: 877,270 | bf16_mfu: 53.89 | total time: 13.71m | eta: 98.0m step 00693/05568 (12.45%) | loss: 2.852956 | lrm: 1.00 | dt: 1197.72ms | tok/sec: 875,478 | bf16_mfu: 53.78 | total time: 13.73m | eta: 98.0m step 00694/05568 (12.46%) | loss: 2.858946 | lrm: 1.00 | dt: 1193.16ms | tok/sec: 878,821 | bf16_mfu: 53.98 | total time: 13.75m | eta: 98.0m step 00695/05568 (12.48%) | loss: 2.848632 | lrm: 1.00 | dt: 1197.95ms | tok/sec: 875,309 | bf16_mfu: 53.77 | total time: 13.77m | eta: 98.0m step 00696/05568 (12.50%) | loss: 2.848992 | lrm: 1.00 | dt: 1193.19ms | tok/sec: 878,801 | bf16_mfu: 53.98 | total time: 13.79m | eta: 97.9m step 00697/05568 (12.52%) | loss: 2.852846 | lrm: 1.00 | dt: 1199.41ms | tok/sec: 874,241 | bf16_mfu: 53.70 | total time: 13.81m | eta: 97.9m step 00698/05568 (12.54%) | loss: 2.858624 | lrm: 1.00 | dt: 1195.48ms | tok/sec: 877,113 | bf16_mfu: 53.88 | total time: 13.83m | eta: 97.9m step 00699/05568 (12.55%) | loss: 2.855006 | lrm: 1.00 | dt: 1194.01ms | tok/sec: 878,197 | bf16_mfu: 53.95 | total time: 13.85m | eta: 97.9m step 00700/05568 (12.57%) | loss: 2.859115 | lrm: 1.00 | dt: 1197.62ms | tok/sec: 875,546 | bf16_mfu: 53.78 | total time: 13.87m | eta: 97.9m step 00701/05568 (12.59%) | loss: 2.855820 | lrm: 1.00 | dt: 1193.94ms | tok/sec: 878,245 | bf16_mfu: 53.95 | total time: 13.89m | eta: 97.8m step 00702/05568 (12.61%) | loss: 2.854027 | lrm: 1.00 | dt: 1200.76ms | tok/sec: 873,258 | bf16_mfu: 53.64 | total time: 13.91m | eta: 97.8m step 00703/05568 (12.63%) | loss: 2.852866 | lrm: 1.00 | dt: 1200.23ms | tok/sec: 873,648 | bf16_mfu: 53.67 | total time: 13.93m | eta: 97.8m step 00704/05568 (12.64%) | loss: 2.861055 | lrm: 1.00 | dt: 1199.44ms | tok/sec: 874,218 | bf16_mfu: 53.70 | total time: 13.95m | eta: 97.8m step 00705/05568 (12.66%) | loss: 2.855473 | lrm: 1.00 | dt: 1200.29ms | tok/sec: 873,602 | bf16_mfu: 53.66 | total time: 13.97m | eta: 97.7m step 00706/05568 (12.68%) | loss: 2.856653 | lrm: 1.00 | dt: 1198.50ms | tok/sec: 874,903 | bf16_mfu: 53.74 | total time: 13.99m | eta: 97.7m step 00707/05568 (12.70%) | loss: 2.861140 | lrm: 1.00 | dt: 1198.38ms | tok/sec: 874,992 | bf16_mfu: 53.75 | total time: 14.01m | eta: 97.7m step 00708/05568 (12.72%) | loss: 2.869149 | lrm: 1.00 | dt: 1192.44ms | tok/sec: 879,355 | bf16_mfu: 54.02 | total time: 14.03m | eta: 97.7m step 00709/05568 (12.73%) | loss: 2.865712 | lrm: 1.00 | dt: 1194.68ms | tok/sec: 877,707 | bf16_mfu: 53.92 | total time: 14.05m | eta: 97.7m step 00710/05568 (12.75%) | loss: 2.869916 | lrm: 1.00 | dt: 1200.09ms | tok/sec: 873,750 | bf16_mfu: 53.67 | total time: 14.07m | eta: 97.6m step 00711/05568 (12.77%) | loss: 2.859142 | lrm: 1.00 | dt: 1191.52ms | tok/sec: 880,031 | bf16_mfu: 54.06 | total time: 14.09m | eta: 97.6m step 00712/05568 (12.79%) | loss: 2.882415 | lrm: 1.00 | dt: 1192.73ms | tok/sec: 879,136 | bf16_mfu: 54.00 | total time: 14.11m | eta: 97.6m step 00713/05568 (12.81%) | loss: 2.880657 | lrm: 1.00 | dt: 1192.43ms | tok/sec: 879,359 | bf16_mfu: 54.02 | total time: 14.13m | eta: 97.6m step 00714/05568 (12.82%) | loss: 2.879707 | lrm: 1.00 | dt: 1195.48ms | tok/sec: 877,116 | bf16_mfu: 53.88 | total time: 14.15m | eta: 97.6m step 00715/05568 (12.84%) | loss: 2.888444 | lrm: 1.00 | dt: 1196.95ms | tok/sec: 876,039 | bf16_mfu: 53.81 | total time: 14.17m | eta: 97.5m step 00716/05568 (12.86%) | loss: 2.883164 | lrm: 1.00 | dt: 1193.15ms | tok/sec: 878,827 | bf16_mfu: 53.98 | total time: 14.19m | eta: 97.5m step 00717/05568 (12.88%) | loss: 2.885008 | lrm: 1.00 | dt: 1194.88ms | tok/sec: 877,557 | bf16_mfu: 53.91 | total time: 14.21m | eta: 97.5m step 00718/05568 (12.90%) | loss: 2.872812 | lrm: 1.00 | dt: 1193.83ms | tok/sec: 878,329 | bf16_mfu: 53.95 | total time: 14.23m | eta: 97.5m step 00719/05568 (12.91%) | loss: 2.872466 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,260 | bf16_mfu: 53.95 | total time: 14.25m | eta: 97.4m step 00720/05568 (12.93%) | loss: 2.876129 | lrm: 1.00 | dt: 1191.14ms | tok/sec: 880,315 | bf16_mfu: 54.08 | total time: 14.27m | eta: 97.4m step 00721/05568 (12.95%) | loss: 2.870364 | lrm: 1.00 | dt: 1197.59ms | tok/sec: 875,575 | bf16_mfu: 53.78 | total time: 14.29m | eta: 97.4m step 00722/05568 (12.97%) | loss: 2.865506 | lrm: 1.00 | dt: 1194.36ms | tok/sec: 877,939 | bf16_mfu: 53.93 | total time: 14.31m | eta: 97.4m step 00723/05568 (12.98%) | loss: 2.863152 | lrm: 1.00 | dt: 1194.61ms | tok/sec: 877,754 | bf16_mfu: 53.92 | total time: 14.33m | eta: 97.4m step 00724/05568 (13.00%) | loss: 2.857847 | lrm: 1.00 | dt: 1199.85ms | tok/sec: 873,923 | bf16_mfu: 53.68 | total time: 14.35m | eta: 97.3m step 00725/05568 (13.02%) | loss: 2.861213 | lrm: 1.00 | dt: 1194.56ms | tok/sec: 877,790 | bf16_mfu: 53.92 | total time: 14.37m | eta: 97.3m step 00726/05568 (13.04%) | loss: 2.869662 | lrm: 1.00 | dt: 1195.22ms | tok/sec: 877,310 | bf16_mfu: 53.89 | total time: 14.39m | eta: 97.3m step 00727/05568 (13.06%) | loss: 2.871589 | lrm: 1.00 | dt: 1199.64ms | tok/sec: 874,076 | bf16_mfu: 53.69 | total time: 14.41m | eta: 97.3m step 00728/05568 (13.07%) | loss: 2.858128 | lrm: 1.00 | dt: 1204.89ms | tok/sec: 870,263 | bf16_mfu: 53.46 | total time: 14.43m | eta: 97.3m step 00729/05568 (13.09%) | loss: 2.858545 | lrm: 1.00 | dt: 1195.54ms | tok/sec: 877,070 | bf16_mfu: 53.88 | total time: 14.45m | eta: 97.2m step 00730/05568 (13.11%) | loss: 2.872266 | lrm: 1.00 | dt: 1196.14ms | tok/sec: 876,635 | bf16_mfu: 53.85 | total time: 14.47m | eta: 97.2m step 00731/05568 (13.13%) | loss: 2.874450 | lrm: 1.00 | dt: 1195.62ms | tok/sec: 877,015 | bf16_mfu: 53.87 | total time: 14.49m | eta: 97.2m step 00732/05568 (13.15%) | loss: 2.872664 | lrm: 1.00 | dt: 1193.47ms | tok/sec: 878,592 | bf16_mfu: 53.97 | total time: 14.51m | eta: 97.2m step 00733/05568 (13.16%) | loss: 2.887955 | lrm: 1.00 | dt: 1197.49ms | tok/sec: 875,648 | bf16_mfu: 53.79 | total time: 14.53m | eta: 97.2m step 00734/05568 (13.18%) | loss: 2.883295 | lrm: 1.00 | dt: 1197.72ms | tok/sec: 875,477 | bf16_mfu: 53.78 | total time: 14.55m | eta: 97.1m step 00735/05568 (13.20%) | loss: 2.876296 | lrm: 1.00 | dt: 1192.82ms | tok/sec: 879,076 | bf16_mfu: 54.00 | total time: 14.57m | eta: 97.1m step 00736/05568 (13.22%) | loss: 2.880725 | lrm: 1.00 | dt: 1192.22ms | tok/sec: 879,515 | bf16_mfu: 54.03 | total time: 14.59m | eta: 97.1m step 00737/05568 (13.24%) | loss: 2.877948 | lrm: 1.00 | dt: 1196.33ms | tok/sec: 876,495 | bf16_mfu: 53.84 | total time: 14.61m | eta: 97.1m step 00738/05568 (13.25%) | loss: 2.876125 | lrm: 1.00 | dt: 1194.01ms | tok/sec: 878,196 | bf16_mfu: 53.95 | total time: 14.63m | eta: 97.0m step 00739/05568 (13.27%) | loss: 2.856866 | lrm: 1.00 | dt: 1203.88ms | tok/sec: 870,998 | bf16_mfu: 53.50 | total time: 14.65m | eta: 97.0m step 00740/05568 (13.29%) | loss: 2.870632 | lrm: 1.00 | dt: 1195.38ms | tok/sec: 877,192 | bf16_mfu: 53.88 | total time: 14.67m | eta: 97.0m step 00741/05568 (13.31%) | loss: 2.872663 | lrm: 1.00 | dt: 1194.41ms | tok/sec: 877,903 | bf16_mfu: 53.93 | total time: 14.69m | eta: 97.0m step 00742/05568 (13.33%) | loss: 2.869597 | lrm: 1.00 | dt: 1198.62ms | tok/sec: 874,816 | bf16_mfu: 53.74 | total time: 14.71m | eta: 97.0m step 00743/05568 (13.34%) | loss: 2.867702 | lrm: 1.00 | dt: 1189.94ms | tok/sec: 881,201 | bf16_mfu: 54.13 | total time: 14.73m | eta: 96.9m step 00744/05568 (13.36%) | loss: 2.867227 | lrm: 1.00 | dt: 1197.44ms | tok/sec: 875,678 | bf16_mfu: 53.79 | total time: 14.75m | eta: 96.9m step 00745/05568 (13.38%) | loss: 2.863428 | lrm: 1.00 | dt: 1199.75ms | tok/sec: 873,993 | bf16_mfu: 53.69 | total time: 14.77m | eta: 96.9m step 00746/05568 (13.40%) | loss: 2.852473 | lrm: 1.00 | dt: 1196.33ms | tok/sec: 876,491 | bf16_mfu: 53.84 | total time: 14.79m | eta: 96.9m step 00747/05568 (13.42%) | loss: 2.855724 | lrm: 1.00 | dt: 1194.00ms | tok/sec: 878,201 | bf16_mfu: 53.95 | total time: 14.81m | eta: 96.9m step 00748/05568 (13.43%) | loss: 2.860970 | lrm: 1.00 | dt: 1191.76ms | tok/sec: 879,857 | bf16_mfu: 54.05 | total time: 14.83m | eta: 96.8m step 00749/05568 (13.45%) | loss: 2.875055 | lrm: 1.00 | dt: 1197.66ms | tok/sec: 875,518 | bf16_mfu: 53.78 | total time: 14.85m | eta: 96.8m step:750/5568 val_bpb:0.866224 val_time:3.89s step 00750/05568 (13.47%) | loss: 2.872977 | lrm: 1.00 | dt: 1193.53ms | tok/sec: 878,549 | bf16_mfu: 53.97 | total time: 14.87m | eta: 96.8m step 00751/05568 (13.49%) | loss: 2.878732 | lrm: 1.00 | dt: 1197.86ms | tok/sec: 875,372 | bf16_mfu: 53.77 | total time: 14.89m | eta: 96.8m step 00752/05568 (13.51%) | loss: 2.867092 | lrm: 1.00 | dt: 1192.50ms | tok/sec: 879,310 | bf16_mfu: 54.01 | total time: 14.91m | eta: 96.7m step 00753/05568 (13.52%) | loss: 2.867397 | lrm: 1.00 | dt: 1192.03ms | tok/sec: 879,657 | bf16_mfu: 54.04 | total time: 14.93m | eta: 96.7m step 00754/05568 (13.54%) | loss: 2.857438 | lrm: 1.00 | dt: 1200.64ms | tok/sec: 873,346 | bf16_mfu: 53.65 | total time: 14.95m | eta: 96.7m step 00755/05568 (13.56%) | loss: 2.859119 | lrm: 1.00 | dt: 1198.52ms | tok/sec: 874,894 | bf16_mfu: 53.74 | total time: 14.97m | eta: 96.7m step 00756/05568 (13.58%) | loss: 2.853553 | lrm: 1.00 | dt: 1198.54ms | tok/sec: 874,876 | bf16_mfu: 53.74 | total time: 14.99m | eta: 96.7m step 00757/05568 (13.60%) | loss: 2.847559 | lrm: 1.00 | dt: 1194.04ms | tok/sec: 878,177 | bf16_mfu: 53.94 | total time: 15.01m | eta: 96.6m step 00758/05568 (13.61%) | loss: 2.837384 | lrm: 1.00 | dt: 1197.07ms | tok/sec: 875,951 | bf16_mfu: 53.81 | total time: 15.03m | eta: 96.6m step 00759/05568 (13.63%) | loss: 2.841757 | lrm: 1.00 | dt: 1192.78ms | tok/sec: 879,101 | bf16_mfu: 54.00 | total time: 15.05m | eta: 96.6m step 00760/05568 (13.65%) | loss: 2.833858 | lrm: 1.00 | dt: 1189.76ms | tok/sec: 881,333 | bf16_mfu: 54.14 | total time: 15.07m | eta: 96.6m step 00761/05568 (13.67%) | loss: 2.830716 | lrm: 1.00 | dt: 1199.18ms | tok/sec: 874,412 | bf16_mfu: 53.71 | total time: 15.09m | eta: 96.6m step 00762/05568 (13.69%) | loss: 2.821053 | lrm: 1.00 | dt: 1200.42ms | tok/sec: 873,504 | bf16_mfu: 53.66 | total time: 15.11m | eta: 96.5m step 00763/05568 (13.70%) | loss: 2.819089 | lrm: 1.00 | dt: 1191.00ms | tok/sec: 880,419 | bf16_mfu: 54.08 | total time: 15.13m | eta: 96.5m step 00764/05568 (13.72%) | loss: 2.819269 | lrm: 1.00 | dt: 1197.35ms | tok/sec: 875,746 | bf16_mfu: 53.80 | total time: 15.15m | eta: 96.5m step 00765/05568 (13.74%) | loss: 2.818637 | lrm: 1.00 | dt: 1196.82ms | tok/sec: 876,136 | bf16_mfu: 53.82 | total time: 15.17m | eta: 96.5m step 00766/05568 (13.76%) | loss: 2.822968 | lrm: 1.00 | dt: 1195.11ms | tok/sec: 877,391 | bf16_mfu: 53.90 | total time: 15.19m | eta: 96.5m step 00767/05568 (13.78%) | loss: 2.819336 | lrm: 1.00 | dt: 1199.47ms | tok/sec: 874,202 | bf16_mfu: 53.70 | total time: 15.21m | eta: 96.4m step 00768/05568 (13.79%) | loss: 2.820737 | lrm: 1.00 | dt: 1192.57ms | tok/sec: 879,254 | bf16_mfu: 54.01 | total time: 15.23m | eta: 96.4m step 00769/05568 (13.81%) | loss: 2.819697 | lrm: 1.00 | dt: 1193.28ms | tok/sec: 878,732 | bf16_mfu: 53.98 | total time: 15.25m | eta: 96.4m step 00770/05568 (13.83%) | loss: 2.822956 | lrm: 1.00 | dt: 1195.15ms | tok/sec: 877,358 | bf16_mfu: 53.89 | total time: 15.27m | eta: 96.4m step 00771/05568 (13.85%) | loss: 2.822907 | lrm: 1.00 | dt: 1197.06ms | tok/sec: 875,958 | bf16_mfu: 53.81 | total time: 15.29m | eta: 96.3m step 00772/05568 (13.86%) | loss: 2.830186 | lrm: 1.00 | dt: 1198.29ms | tok/sec: 875,058 | bf16_mfu: 53.75 | total time: 15.31m | eta: 96.3m step 00773/05568 (13.88%) | loss: 2.828820 | lrm: 1.00 | dt: 1190.09ms | tok/sec: 881,089 | bf16_mfu: 54.12 | total time: 15.33m | eta: 96.3m step 00774/05568 (13.90%) | loss: 2.832043 | lrm: 1.00 | dt: 1199.18ms | tok/sec: 874,408 | bf16_mfu: 53.71 | total time: 15.35m | eta: 96.3m step 00775/05568 (13.92%) | loss: 2.834126 | lrm: 1.00 | dt: 1190.68ms | tok/sec: 880,652 | bf16_mfu: 54.10 | total time: 15.37m | eta: 96.3m step 00776/05568 (13.94%) | loss: 2.850240 | lrm: 1.00 | dt: 1195.88ms | tok/sec: 876,822 | bf16_mfu: 53.86 | total time: 15.39m | eta: 96.2m step 00777/05568 (13.95%) | loss: 2.834420 | lrm: 1.00 | dt: 1197.89ms | tok/sec: 875,350 | bf16_mfu: 53.77 | total time: 15.41m | eta: 96.2m step 00778/05568 (13.97%) | loss: 2.848078 | lrm: 1.00 | dt: 1193.42ms | tok/sec: 878,627 | bf16_mfu: 53.97 | total time: 15.43m | eta: 96.2m step 00779/05568 (13.99%) | loss: 2.846061 | lrm: 1.00 | dt: 1193.77ms | tok/sec: 878,373 | bf16_mfu: 53.96 | total time: 15.45m | eta: 96.2m step 00780/05568 (14.01%) | loss: 2.846593 | lrm: 1.00 | dt: 1190.94ms | tok/sec: 880,459 | bf16_mfu: 54.08 | total time: 15.47m | eta: 96.2m step 00781/05568 (14.03%) | loss: 2.845028 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,640 | bf16_mfu: 53.91 | total time: 15.49m | eta: 96.1m step 00782/05568 (14.04%) | loss: 2.846831 | lrm: 1.00 | dt: 1196.93ms | tok/sec: 876,054 | bf16_mfu: 53.81 | total time: 15.51m | eta: 96.1m step 00783/05568 (14.06%) | loss: 2.840185 | lrm: 1.00 | dt: 1196.96ms | tok/sec: 876,036 | bf16_mfu: 53.81 | total time: 15.53m | eta: 96.1m step 00784/05568 (14.08%) | loss: 2.830477 | lrm: 1.00 | dt: 1196.72ms | tok/sec: 876,205 | bf16_mfu: 53.82 | total time: 15.55m | eta: 96.1m step 00785/05568 (14.10%) | loss: 2.834469 | lrm: 1.00 | dt: 1199.75ms | tok/sec: 873,991 | bf16_mfu: 53.69 | total time: 15.57m | eta: 96.1m step 00786/05568 (14.12%) | loss: 2.827768 | lrm: 1.00 | dt: 1200.78ms | tok/sec: 873,246 | bf16_mfu: 53.64 | total time: 15.59m | eta: 96.0m step 00787/05568 (14.13%) | loss: 2.827235 | lrm: 1.00 | dt: 1193.67ms | tok/sec: 878,448 | bf16_mfu: 53.96 | total time: 15.61m | eta: 96.0m step 00788/05568 (14.15%) | loss: 2.826297 | lrm: 1.00 | dt: 1192.77ms | tok/sec: 879,112 | bf16_mfu: 54.00 | total time: 15.63m | eta: 96.0m step 00789/05568 (14.17%) | loss: 2.819800 | lrm: 1.00 | dt: 1194.89ms | tok/sec: 877,553 | bf16_mfu: 53.91 | total time: 15.65m | eta: 96.0m step 00790/05568 (14.19%) | loss: 2.811661 | lrm: 1.00 | dt: 1195.57ms | tok/sec: 877,051 | bf16_mfu: 53.88 | total time: 15.67m | eta: 95.9m step 00791/05568 (14.21%) | loss: 2.823299 | lrm: 1.00 | dt: 1194.65ms | tok/sec: 877,728 | bf16_mfu: 53.92 | total time: 15.69m | eta: 95.9m step 00792/05568 (14.22%) | loss: 2.822993 | lrm: 1.00 | dt: 1193.37ms | tok/sec: 878,667 | bf16_mfu: 53.97 | total time: 15.71m | eta: 95.9m step 00793/05568 (14.24%) | loss: 2.828488 | lrm: 1.00 | dt: 1201.47ms | tok/sec: 872,743 | bf16_mfu: 53.61 | total time: 15.73m | eta: 95.9m step 00794/05568 (14.26%) | loss: 2.825129 | lrm: 1.00 | dt: 1193.36ms | tok/sec: 878,677 | bf16_mfu: 53.98 | total time: 15.75m | eta: 95.9m step 00795/05568 (14.28%) | loss: 2.824451 | lrm: 1.00 | dt: 1195.66ms | tok/sec: 876,987 | bf16_mfu: 53.87 | total time: 15.77m | eta: 95.8m step 00796/05568 (14.30%) | loss: 2.832473 | lrm: 1.00 | dt: 1192.68ms | tok/sec: 879,176 | bf16_mfu: 54.01 | total time: 15.79m | eta: 95.8m step 00797/05568 (14.31%) | loss: 2.845967 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,636 | bf16_mfu: 53.91 | total time: 15.81m | eta: 95.8m step 00798/05568 (14.33%) | loss: 2.851170 | lrm: 1.00 | dt: 1194.21ms | tok/sec: 878,048 | bf16_mfu: 53.94 | total time: 15.83m | eta: 95.8m step 00799/05568 (14.35%) | loss: 2.850693 | lrm: 1.00 | dt: 1200.62ms | tok/sec: 873,359 | bf16_mfu: 53.65 | total time: 15.85m | eta: 95.8m step 00800/05568 (14.37%) | loss: 2.841271 | lrm: 1.00 | dt: 1201.01ms | tok/sec: 873,082 | bf16_mfu: 53.63 | total time: 15.87m | eta: 95.7m step 00801/05568 (14.39%) | loss: 2.840756 | lrm: 1.00 | dt: 1193.01ms | tok/sec: 878,933 | bf16_mfu: 53.99 | total time: 15.89m | eta: 95.7m step 00802/05568 (14.40%) | loss: 2.843785 | lrm: 1.00 | dt: 1192.26ms | tok/sec: 879,488 | bf16_mfu: 54.02 | total time: 15.91m | eta: 95.7m step 00803/05568 (14.42%) | loss: 2.833082 | lrm: 1.00 | dt: 1192.98ms | tok/sec: 878,956 | bf16_mfu: 53.99 | total time: 15.93m | eta: 95.7m step 00804/05568 (14.44%) | loss: 2.825082 | lrm: 1.00 | dt: 1192.02ms | tok/sec: 879,661 | bf16_mfu: 54.04 | total time: 15.95m | eta: 95.7m step 00805/05568 (14.46%) | loss: 2.828214 | lrm: 1.00 | dt: 1189.13ms | tok/sec: 881,798 | bf16_mfu: 54.17 | total time: 15.96m | eta: 95.6m step 00806/05568 (14.48%) | loss: 2.817065 | lrm: 1.00 | dt: 1199.27ms | tok/sec: 874,346 | bf16_mfu: 53.71 | total time: 15.98m | eta: 95.6m step 00807/05568 (14.49%) | loss: 2.813992 | lrm: 1.00 | dt: 1197.20ms | tok/sec: 875,857 | bf16_mfu: 53.80 | total time: 16.00m | eta: 95.6m step 00808/05568 (14.51%) | loss: 2.816486 | lrm: 1.00 | dt: 1193.46ms | tok/sec: 878,598 | bf16_mfu: 53.97 | total time: 16.02m | eta: 95.6m step 00809/05568 (14.53%) | loss: 2.824356 | lrm: 1.00 | dt: 1195.80ms | tok/sec: 876,884 | bf16_mfu: 53.87 | total time: 16.04m | eta: 95.5m step 00810/05568 (14.55%) | loss: 2.829136 | lrm: 1.00 | dt: 1193.69ms | tok/sec: 878,430 | bf16_mfu: 53.96 | total time: 16.06m | eta: 95.5m step 00811/05568 (14.57%) | loss: 2.830951 | lrm: 1.00 | dt: 1193.07ms | tok/sec: 878,889 | bf16_mfu: 53.99 | total time: 16.08m | eta: 95.5m step 00812/05568 (14.58%) | loss: 2.834948 | lrm: 1.00 | dt: 1198.43ms | tok/sec: 874,959 | bf16_mfu: 53.75 | total time: 16.10m | eta: 95.5m step 00813/05568 (14.60%) | loss: 2.828667 | lrm: 1.00 | dt: 1197.73ms | tok/sec: 875,470 | bf16_mfu: 53.78 | total time: 16.12m | eta: 95.5m step 00814/05568 (14.62%) | loss: 2.824044 | lrm: 1.00 | dt: 1190.39ms | tok/sec: 880,866 | bf16_mfu: 54.11 | total time: 16.14m | eta: 95.4m step 00815/05568 (14.64%) | loss: 2.831495 | lrm: 1.00 | dt: 1194.18ms | tok/sec: 878,073 | bf16_mfu: 53.94 | total time: 16.16m | eta: 95.4m step 00816/05568 (14.66%) | loss: 2.829354 | lrm: 1.00 | dt: 1196.35ms | tok/sec: 876,476 | bf16_mfu: 53.84 | total time: 16.18m | eta: 95.4m step 00817/05568 (14.67%) | loss: 2.832092 | lrm: 1.00 | dt: 1192.17ms | tok/sec: 879,554 | bf16_mfu: 54.03 | total time: 16.20m | eta: 95.4m step 00818/05568 (14.69%) | loss: 2.847442 | lrm: 1.00 | dt: 1196.76ms | tok/sec: 876,177 | bf16_mfu: 53.82 | total time: 16.22m | eta: 95.4m step 00819/05568 (14.71%) | loss: 2.850402 | lrm: 1.00 | dt: 1196.08ms | tok/sec: 876,673 | bf16_mfu: 53.85 | total time: 16.24m | eta: 95.3m step 00820/05568 (14.73%) | loss: 2.869185 | lrm: 1.00 | dt: 1195.79ms | tok/sec: 876,891 | bf16_mfu: 53.87 | total time: 16.26m | eta: 95.3m step 00821/05568 (14.74%) | loss: 2.856069 | lrm: 1.00 | dt: 1192.00ms | tok/sec: 879,675 | bf16_mfu: 54.04 | total time: 16.28m | eta: 95.3m step 00822/05568 (14.76%) | loss: 2.860524 | lrm: 1.00 | dt: 1196.54ms | tok/sec: 876,338 | bf16_mfu: 53.83 | total time: 16.30m | eta: 95.3m step 00823/05568 (14.78%) | loss: 2.855787 | lrm: 1.00 | dt: 1194.19ms | tok/sec: 878,065 | bf16_mfu: 53.94 | total time: 16.32m | eta: 95.3m step 00824/05568 (14.80%) | loss: 2.859125 | lrm: 1.00 | dt: 1197.45ms | tok/sec: 875,672 | bf16_mfu: 53.79 | total time: 16.34m | eta: 95.2m step 00825/05568 (14.82%) | loss: 2.863311 | lrm: 1.00 | dt: 1194.10ms | tok/sec: 878,134 | bf16_mfu: 53.94 | total time: 16.36m | eta: 95.2m step 00826/05568 (14.83%) | loss: 2.861657 | lrm: 1.00 | dt: 1196.25ms | tok/sec: 876,550 | bf16_mfu: 53.84 | total time: 16.38m | eta: 95.2m step 00827/05568 (14.85%) | loss: 2.844881 | lrm: 1.00 | dt: 1194.63ms | tok/sec: 877,738 | bf16_mfu: 53.92 | total time: 16.40m | eta: 95.2m step 00828/05568 (14.87%) | loss: 2.848671 | lrm: 1.00 | dt: 1195.37ms | tok/sec: 877,197 | bf16_mfu: 53.88 | total time: 16.42m | eta: 95.1m step 00829/05568 (14.89%) | loss: 2.846206 | lrm: 1.00 | dt: 1190.63ms | tok/sec: 880,688 | bf16_mfu: 54.10 | total time: 16.44m | eta: 95.1m step 00830/05568 (14.91%) | loss: 2.841750 | lrm: 1.00 | dt: 1194.01ms | tok/sec: 878,199 | bf16_mfu: 53.95 | total time: 16.46m | eta: 95.1m step 00831/05568 (14.92%) | loss: 2.848233 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,395 | bf16_mfu: 54.02 | total time: 16.48m | eta: 95.1m step 00832/05568 (14.94%) | loss: 2.841944 | lrm: 1.00 | dt: 1191.27ms | tok/sec: 880,217 | bf16_mfu: 54.07 | total time: 16.50m | eta: 95.1m step 00833/05568 (14.96%) | loss: 2.845394 | lrm: 1.00 | dt: 1196.34ms | tok/sec: 876,484 | bf16_mfu: 53.84 | total time: 16.52m | eta: 95.0m step 00834/05568 (14.98%) | loss: 2.825140 | lrm: 1.00 | dt: 1197.37ms | tok/sec: 875,729 | bf16_mfu: 53.79 | total time: 16.54m | eta: 95.0m step 00835/05568 (15.00%) | loss: 2.822048 | lrm: 1.00 | dt: 1195.30ms | tok/sec: 877,246 | bf16_mfu: 53.89 | total time: 16.56m | eta: 95.0m step 00836/05568 (15.01%) | loss: 2.810018 | lrm: 1.00 | dt: 1196.81ms | tok/sec: 876,142 | bf16_mfu: 53.82 | total time: 16.58m | eta: 95.0m step 00837/05568 (15.03%) | loss: 2.811759 | lrm: 1.00 | dt: 1197.56ms | tok/sec: 875,592 | bf16_mfu: 53.79 | total time: 16.60m | eta: 95.0m step 00838/05568 (15.05%) | loss: 2.814367 | lrm: 1.00 | dt: 1195.55ms | tok/sec: 877,063 | bf16_mfu: 53.88 | total time: 16.62m | eta: 94.9m step 00839/05568 (15.07%) | loss: 2.817578 | lrm: 1.00 | dt: 1193.31ms | tok/sec: 878,714 | bf16_mfu: 53.98 | total time: 16.64m | eta: 94.9m step 00840/05568 (15.09%) | loss: 2.819516 | lrm: 1.00 | dt: 1190.46ms | tok/sec: 880,818 | bf16_mfu: 54.11 | total time: 16.66m | eta: 94.9m step 00841/05568 (15.10%) | loss: 2.811547 | lrm: 1.00 | dt: 1199.25ms | tok/sec: 874,357 | bf16_mfu: 53.71 | total time: 16.68m | eta: 94.9m step 00842/05568 (15.12%) | loss: 2.821249 | lrm: 1.00 | dt: 1190.75ms | tok/sec: 880,600 | bf16_mfu: 54.09 | total time: 16.70m | eta: 94.9m step 00843/05568 (15.14%) | loss: 2.819403 | lrm: 1.00 | dt: 1197.05ms | tok/sec: 875,967 | bf16_mfu: 53.81 | total time: 16.72m | eta: 94.8m step 00844/05568 (15.16%) | loss: 2.823906 | lrm: 1.00 | dt: 1193.75ms | tok/sec: 878,387 | bf16_mfu: 53.96 | total time: 16.74m | eta: 94.8m step 00845/05568 (15.18%) | loss: 2.833485 | lrm: 1.00 | dt: 1190.02ms | tok/sec: 881,141 | bf16_mfu: 54.13 | total time: 16.76m | eta: 94.8m step 00846/05568 (15.19%) | loss: 2.836141 | lrm: 1.00 | dt: 1195.31ms | tok/sec: 877,244 | bf16_mfu: 53.89 | total time: 16.78m | eta: 94.8m step 00847/05568 (15.21%) | loss: 2.829961 | lrm: 1.00 | dt: 1195.15ms | tok/sec: 877,361 | bf16_mfu: 53.89 | total time: 16.80m | eta: 94.7m step 00848/05568 (15.23%) | loss: 2.829966 | lrm: 1.00 | dt: 1189.72ms | tok/sec: 881,363 | bf16_mfu: 54.14 | total time: 16.82m | eta: 94.7m step 00849/05568 (15.25%) | loss: 2.826046 | lrm: 1.00 | dt: 1189.45ms | tok/sec: 881,565 | bf16_mfu: 54.15 | total time: 16.84m | eta: 94.7m step 00850/05568 (15.27%) | loss: 2.836241 | lrm: 1.00 | dt: 1188.91ms | tok/sec: 881,966 | bf16_mfu: 54.18 | total time: 16.86m | eta: 94.7m step 00851/05568 (15.28%) | loss: 2.828568 | lrm: 1.00 | dt: 1194.09ms | tok/sec: 878,139 | bf16_mfu: 53.94 | total time: 16.88m | eta: 94.7m step 00852/05568 (15.30%) | loss: 2.831933 | lrm: 1.00 | dt: 1199.37ms | tok/sec: 874,268 | bf16_mfu: 53.70 | total time: 16.90m | eta: 94.6m step 00853/05568 (15.32%) | loss: 2.833714 | lrm: 1.00 | dt: 1195.86ms | tok/sec: 876,841 | bf16_mfu: 53.86 | total time: 16.92m | eta: 94.6m step 00854/05568 (15.34%) | loss: 2.826869 | lrm: 1.00 | dt: 1196.57ms | tok/sec: 876,318 | bf16_mfu: 53.83 | total time: 16.94m | eta: 94.6m step 00855/05568 (15.36%) | loss: 2.821362 | lrm: 1.00 | dt: 1198.41ms | tok/sec: 874,975 | bf16_mfu: 53.75 | total time: 16.96m | eta: 94.6m step 00856/05568 (15.37%) | loss: 2.804243 | lrm: 1.00 | dt: 1194.84ms | tok/sec: 877,585 | bf16_mfu: 53.91 | total time: 16.98m | eta: 94.6m step 00857/05568 (15.39%) | loss: 2.792915 | lrm: 1.00 | dt: 1194.07ms | tok/sec: 878,154 | bf16_mfu: 53.94 | total time: 17.00m | eta: 94.5m step 00858/05568 (15.41%) | loss: 2.794133 | lrm: 1.00 | dt: 1201.48ms | tok/sec: 872,734 | bf16_mfu: 53.61 | total time: 17.02m | eta: 94.5m step 00859/05568 (15.43%) | loss: 2.810624 | lrm: 1.00 | dt: 1192.77ms | tok/sec: 879,107 | bf16_mfu: 54.00 | total time: 17.04m | eta: 94.5m step 00860/05568 (15.45%) | loss: 2.798273 | lrm: 1.00 | dt: 1191.50ms | tok/sec: 880,045 | bf16_mfu: 54.06 | total time: 17.06m | eta: 94.5m step 00861/05568 (15.46%) | loss: 2.804606 | lrm: 1.00 | dt: 1192.96ms | tok/sec: 878,966 | bf16_mfu: 53.99 | total time: 17.08m | eta: 94.5m step 00862/05568 (15.48%) | loss: 2.794315 | lrm: 1.00 | dt: 1199.30ms | tok/sec: 874,321 | bf16_mfu: 53.71 | total time: 17.10m | eta: 94.4m step 00863/05568 (15.50%) | loss: 2.796582 | lrm: 1.00 | dt: 1191.70ms | tok/sec: 879,896 | bf16_mfu: 54.05 | total time: 17.12m | eta: 94.4m step 00864/05568 (15.52%) | loss: 2.792097 | lrm: 1.00 | dt: 1201.07ms | tok/sec: 873,033 | bf16_mfu: 53.63 | total time: 17.14m | eta: 94.4m step 00865/05568 (15.54%) | loss: 2.798263 | lrm: 1.00 | dt: 1191.79ms | tok/sec: 879,831 | bf16_mfu: 54.05 | total time: 17.16m | eta: 94.4m step 00866/05568 (15.55%) | loss: 2.803815 | lrm: 1.00 | dt: 1202.77ms | tok/sec: 871,801 | bf16_mfu: 53.55 | total time: 17.18m | eta: 94.3m step 00867/05568 (15.57%) | loss: 2.802370 | lrm: 1.00 | dt: 1196.61ms | tok/sec: 876,288 | bf16_mfu: 53.83 | total time: 17.20m | eta: 94.3m step 00868/05568 (15.59%) | loss: 2.802574 | lrm: 1.00 | dt: 1195.18ms | tok/sec: 877,339 | bf16_mfu: 53.89 | total time: 17.22m | eta: 94.3m step 00869/05568 (15.61%) | loss: 2.807383 | lrm: 1.00 | dt: 1194.65ms | tok/sec: 877,724 | bf16_mfu: 53.92 | total time: 17.24m | eta: 94.3m step 00870/05568 (15.62%) | loss: 2.820746 | lrm: 1.00 | dt: 1193.09ms | tok/sec: 878,876 | bf16_mfu: 53.99 | total time: 17.26m | eta: 94.3m step 00871/05568 (15.64%) | loss: 2.819908 | lrm: 1.00 | dt: 1195.57ms | tok/sec: 877,049 | bf16_mfu: 53.88 | total time: 17.28m | eta: 94.2m step 00872/05568 (15.66%) | loss: 2.809872 | lrm: 1.00 | dt: 1197.25ms | tok/sec: 875,817 | bf16_mfu: 53.80 | total time: 17.30m | eta: 94.2m step 00873/05568 (15.68%) | loss: 2.812812 | lrm: 1.00 | dt: 1193.88ms | tok/sec: 878,293 | bf16_mfu: 53.95 | total time: 17.32m | eta: 94.2m step 00874/05568 (15.70%) | loss: 2.815536 | lrm: 1.00 | dt: 1195.50ms | tok/sec: 877,105 | bf16_mfu: 53.88 | total time: 17.34m | eta: 94.2m step 00875/05568 (15.71%) | loss: 2.807598 | lrm: 1.00 | dt: 1197.84ms | tok/sec: 875,388 | bf16_mfu: 53.77 | total time: 17.36m | eta: 94.2m step 00876/05568 (15.73%) | loss: 2.801719 | lrm: 1.00 | dt: 1190.67ms | tok/sec: 880,657 | bf16_mfu: 54.10 | total time: 17.38m | eta: 94.1m step 00877/05568 (15.75%) | loss: 2.799518 | lrm: 1.00 | dt: 1187.00ms | tok/sec: 883,381 | bf16_mfu: 54.26 | total time: 17.40m | eta: 94.1m step 00878/05568 (15.77%) | loss: 2.801836 | lrm: 1.00 | dt: 1194.29ms | tok/sec: 877,991 | bf16_mfu: 53.93 | total time: 17.42m | eta: 94.1m step 00879/05568 (15.79%) | loss: 2.802901 | lrm: 1.00 | dt: 1194.89ms | tok/sec: 877,549 | bf16_mfu: 53.91 | total time: 17.44m | eta: 94.1m step 00880/05568 (15.80%) | loss: 2.798369 | lrm: 1.00 | dt: 1193.53ms | tok/sec: 878,550 | bf16_mfu: 53.97 | total time: 17.46m | eta: 94.1m step 00881/05568 (15.82%) | loss: 2.806605 | lrm: 1.00 | dt: 1195.11ms | tok/sec: 877,388 | bf16_mfu: 53.90 | total time: 17.48m | eta: 94.0m step 00882/05568 (15.84%) | loss: 2.800487 | lrm: 1.00 | dt: 1192.81ms | tok/sec: 879,081 | bf16_mfu: 54.00 | total time: 17.50m | eta: 94.0m step 00883/05568 (15.86%) | loss: 2.799879 | lrm: 1.00 | dt: 1189.43ms | tok/sec: 881,575 | bf16_mfu: 54.15 | total time: 17.52m | eta: 94.0m step 00884/05568 (15.88%) | loss: 2.808361 | lrm: 1.00 | dt: 1193.55ms | tok/sec: 878,532 | bf16_mfu: 53.97 | total time: 17.54m | eta: 94.0m step 00885/05568 (15.89%) | loss: 2.812533 | lrm: 1.00 | dt: 1191.71ms | tok/sec: 879,889 | bf16_mfu: 54.05 | total time: 17.56m | eta: 93.9m step 00886/05568 (15.91%) | loss: 2.828940 | lrm: 1.00 | dt: 1198.87ms | tok/sec: 874,633 | bf16_mfu: 53.73 | total time: 17.58m | eta: 93.9m step 00887/05568 (15.93%) | loss: 2.830399 | lrm: 1.00 | dt: 1192.71ms | tok/sec: 879,155 | bf16_mfu: 54.00 | total time: 17.60m | eta: 93.9m step 00888/05568 (15.95%) | loss: 2.827326 | lrm: 1.00 | dt: 1195.58ms | tok/sec: 877,043 | bf16_mfu: 53.87 | total time: 17.62m | eta: 93.9m step 00889/05568 (15.97%) | loss: 2.816066 | lrm: 1.00 | dt: 1197.37ms | tok/sec: 875,733 | bf16_mfu: 53.79 | total time: 17.64m | eta: 93.9m step 00890/05568 (15.98%) | loss: 2.809335 | lrm: 1.00 | dt: 1195.71ms | tok/sec: 876,949 | bf16_mfu: 53.87 | total time: 17.66m | eta: 93.8m step 00891/05568 (16.00%) | loss: 2.817817 | lrm: 1.00 | dt: 1193.13ms | tok/sec: 878,842 | bf16_mfu: 53.99 | total time: 17.68m | eta: 93.8m step 00892/05568 (16.02%) | loss: 2.817366 | lrm: 1.00 | dt: 1192.44ms | tok/sec: 879,353 | bf16_mfu: 54.02 | total time: 17.70m | eta: 93.8m step 00893/05568 (16.04%) | loss: 2.812922 | lrm: 1.00 | dt: 1199.40ms | tok/sec: 874,249 | bf16_mfu: 53.70 | total time: 17.72m | eta: 93.8m step 00894/05568 (16.06%) | loss: 2.813700 | lrm: 1.00 | dt: 1196.14ms | tok/sec: 876,629 | bf16_mfu: 53.85 | total time: 17.74m | eta: 93.8m step 00895/05568 (16.07%) | loss: 2.808585 | lrm: 1.00 | dt: 1198.49ms | tok/sec: 874,915 | bf16_mfu: 53.74 | total time: 17.76m | eta: 93.7m step 00896/05568 (16.09%) | loss: 2.808677 | lrm: 1.00 | dt: 1191.09ms | tok/sec: 880,347 | bf16_mfu: 54.08 | total time: 17.78m | eta: 93.7m step 00897/05568 (16.11%) | loss: 2.809730 | lrm: 1.00 | dt: 1191.68ms | tok/sec: 879,912 | bf16_mfu: 54.05 | total time: 17.80m | eta: 93.7m step 00898/05568 (16.13%) | loss: 2.813171 | lrm: 1.00 | dt: 1196.87ms | tok/sec: 876,096 | bf16_mfu: 53.82 | total time: 17.82m | eta: 93.7m step 00899/05568 (16.15%) | loss: 2.810497 | lrm: 1.00 | dt: 1192.90ms | tok/sec: 879,013 | bf16_mfu: 54.00 | total time: 17.84m | eta: 93.7m step 00900/05568 (16.16%) | loss: 2.824565 | lrm: 1.00 | dt: 1195.36ms | tok/sec: 877,201 | bf16_mfu: 53.88 | total time: 17.86m | eta: 93.6m step 00901/05568 (16.18%) | loss: 2.825532 | lrm: 1.00 | dt: 1196.54ms | tok/sec: 876,338 | bf16_mfu: 53.83 | total time: 17.88m | eta: 93.6m step 00902/05568 (16.20%) | loss: 2.813760 | lrm: 1.00 | dt: 1186.94ms | tok/sec: 883,429 | bf16_mfu: 54.27 | total time: 17.90m | eta: 93.6m step 00903/05568 (16.22%) | loss: 2.803823 | lrm: 1.00 | dt: 1195.61ms | tok/sec: 877,022 | bf16_mfu: 53.87 | total time: 17.92m | eta: 93.6m step 00904/05568 (16.24%) | loss: 2.809037 | lrm: 1.00 | dt: 1193.48ms | tok/sec: 878,585 | bf16_mfu: 53.97 | total time: 17.94m | eta: 93.6m step 00905/05568 (16.25%) | loss: 2.803770 | lrm: 1.00 | dt: 1192.71ms | tok/sec: 879,153 | bf16_mfu: 54.00 | total time: 17.96m | eta: 93.5m step 00906/05568 (16.27%) | loss: 2.796227 | lrm: 1.00 | dt: 1196.64ms | tok/sec: 876,266 | bf16_mfu: 53.83 | total time: 17.98m | eta: 93.5m step 00907/05568 (16.29%) | loss: 2.803354 | lrm: 1.00 | dt: 1198.05ms | tok/sec: 875,237 | bf16_mfu: 53.76 | total time: 18.00m | eta: 93.5m step 00908/05568 (16.31%) | loss: 2.803552 | lrm: 1.00 | dt: 1191.24ms | tok/sec: 880,240 | bf16_mfu: 54.07 | total time: 18.02m | eta: 93.5m step 00909/05568 (16.33%) | loss: 2.800125 | lrm: 1.00 | dt: 1191.64ms | tok/sec: 879,942 | bf16_mfu: 54.05 | total time: 18.04m | eta: 93.4m step 00910/05568 (16.34%) | loss: 2.813572 | lrm: 1.00 | dt: 1190.09ms | tok/sec: 881,092 | bf16_mfu: 54.12 | total time: 18.06m | eta: 93.4m step 00911/05568 (16.36%) | loss: 2.818053 | lrm: 1.00 | dt: 1194.44ms | tok/sec: 877,884 | bf16_mfu: 53.93 | total time: 18.08m | eta: 93.4m step 00912/05568 (16.38%) | loss: 2.820685 | lrm: 1.00 | dt: 1196.37ms | tok/sec: 876,462 | bf16_mfu: 53.84 | total time: 18.10m | eta: 93.4m step 00913/05568 (16.40%) | loss: 2.821350 | lrm: 1.00 | dt: 1193.53ms | tok/sec: 878,546 | bf16_mfu: 53.97 | total time: 18.12m | eta: 93.4m step 00914/05568 (16.42%) | loss: 2.811241 | lrm: 1.00 | dt: 1200.09ms | tok/sec: 873,745 | bf16_mfu: 53.67 | total time: 18.14m | eta: 93.3m step 00915/05568 (16.43%) | loss: 2.806311 | lrm: 1.00 | dt: 1200.23ms | tok/sec: 873,648 | bf16_mfu: 53.67 | total time: 18.16m | eta: 93.3m step 00916/05568 (16.45%) | loss: 2.810210 | lrm: 1.00 | dt: 1194.72ms | tok/sec: 877,671 | bf16_mfu: 53.91 | total time: 18.18m | eta: 93.3m step 00917/05568 (16.47%) | loss: 2.808084 | lrm: 1.00 | dt: 1196.22ms | tok/sec: 876,572 | bf16_mfu: 53.85 | total time: 18.20m | eta: 93.3m step 00918/05568 (16.49%) | loss: 2.813417 | lrm: 1.00 | dt: 1194.48ms | tok/sec: 877,852 | bf16_mfu: 53.92 | total time: 18.21m | eta: 93.3m step 00919/05568 (16.51%) | loss: 2.812474 | lrm: 1.00 | dt: 1196.16ms | tok/sec: 876,614 | bf16_mfu: 53.85 | total time: 18.23m | eta: 93.2m step 00920/05568 (16.52%) | loss: 2.802184 | lrm: 1.00 | dt: 1191.64ms | tok/sec: 879,940 | bf16_mfu: 54.05 | total time: 18.25m | eta: 93.2m step 00921/05568 (16.54%) | loss: 2.798125 | lrm: 1.00 | dt: 1191.20ms | tok/sec: 880,266 | bf16_mfu: 54.07 | total time: 18.27m | eta: 93.2m step 00922/05568 (16.56%) | loss: 2.795365 | lrm: 1.00 | dt: 1197.62ms | tok/sec: 875,546 | bf16_mfu: 53.78 | total time: 18.29m | eta: 93.2m step 00923/05568 (16.58%) | loss: 2.798777 | lrm: 1.00 | dt: 1199.90ms | tok/sec: 873,886 | bf16_mfu: 53.68 | total time: 18.31m | eta: 93.2m step 00924/05568 (16.59%) | loss: 2.799105 | lrm: 1.00 | dt: 1194.74ms | tok/sec: 877,659 | bf16_mfu: 53.91 | total time: 18.33m | eta: 93.1m step 00925/05568 (16.61%) | loss: 2.797391 | lrm: 1.00 | dt: 1197.93ms | tok/sec: 875,325 | bf16_mfu: 53.77 | total time: 18.35m | eta: 93.1m step 00926/05568 (16.63%) | loss: 2.788238 | lrm: 1.00 | dt: 1194.39ms | tok/sec: 877,916 | bf16_mfu: 53.93 | total time: 18.37m | eta: 93.1m step 00927/05568 (16.65%) | loss: 2.781937 | lrm: 1.00 | dt: 1191.21ms | tok/sec: 880,258 | bf16_mfu: 54.07 | total time: 18.39m | eta: 93.1m step 00928/05568 (16.67%) | loss: 2.786509 | lrm: 1.00 | dt: 1193.20ms | tok/sec: 878,794 | bf16_mfu: 53.98 | total time: 18.41m | eta: 93.1m step 00929/05568 (16.68%) | loss: 2.772566 | lrm: 1.00 | dt: 1195.49ms | tok/sec: 877,113 | bf16_mfu: 53.88 | total time: 18.43m | eta: 93.0m step 00930/05568 (16.70%) | loss: 2.780321 | lrm: 1.00 | dt: 1193.10ms | tok/sec: 878,864 | bf16_mfu: 53.99 | total time: 18.45m | eta: 93.0m step 00931/05568 (16.72%) | loss: 2.780479 | lrm: 1.00 | dt: 1195.44ms | tok/sec: 877,150 | bf16_mfu: 53.88 | total time: 18.47m | eta: 93.0m step 00932/05568 (16.74%) | loss: 2.776480 | lrm: 1.00 | dt: 1192.24ms | tok/sec: 879,501 | bf16_mfu: 54.03 | total time: 18.49m | eta: 93.0m step 00933/05568 (16.76%) | loss: 2.780528 | lrm: 1.00 | dt: 1193.21ms | tok/sec: 878,783 | bf16_mfu: 53.98 | total time: 18.51m | eta: 92.9m step 00934/05568 (16.77%) | loss: 2.778410 | lrm: 1.00 | dt: 1199.65ms | tok/sec: 874,067 | bf16_mfu: 53.69 | total time: 18.53m | eta: 92.9m step 00935/05568 (16.79%) | loss: 2.776758 | lrm: 1.00 | dt: 1195.77ms | tok/sec: 876,901 | bf16_mfu: 53.87 | total time: 18.55m | eta: 92.9m step 00936/05568 (16.81%) | loss: 2.777420 | lrm: 1.00 | dt: 1196.71ms | tok/sec: 876,212 | bf16_mfu: 53.82 | total time: 18.57m | eta: 92.9m step 00937/05568 (16.83%) | loss: 2.789232 | lrm: 1.00 | dt: 1195.36ms | tok/sec: 877,208 | bf16_mfu: 53.88 | total time: 18.59m | eta: 92.9m step 00938/05568 (16.85%) | loss: 2.776693 | lrm: 1.00 | dt: 1196.57ms | tok/sec: 876,320 | bf16_mfu: 53.83 | total time: 18.61m | eta: 92.8m step 00939/05568 (16.86%) | loss: 2.772335 | lrm: 1.00 | dt: 1198.11ms | tok/sec: 875,192 | bf16_mfu: 53.76 | total time: 18.63m | eta: 92.8m step 00940/05568 (16.88%) | loss: 2.774747 | lrm: 1.00 | dt: 1192.32ms | tok/sec: 879,442 | bf16_mfu: 54.02 | total time: 18.65m | eta: 92.8m step 00941/05568 (16.90%) | loss: 2.772936 | lrm: 1.00 | dt: 1198.50ms | tok/sec: 874,906 | bf16_mfu: 53.74 | total time: 18.67m | eta: 92.8m step 00942/05568 (16.92%) | loss: 2.787626 | lrm: 1.00 | dt: 1199.02ms | tok/sec: 874,525 | bf16_mfu: 53.72 | total time: 18.69m | eta: 92.8m step 00943/05568 (16.94%) | loss: 2.783997 | lrm: 1.00 | dt: 1195.65ms | tok/sec: 876,992 | bf16_mfu: 53.87 | total time: 18.71m | eta: 92.7m step 00944/05568 (16.95%) | loss: 2.774858 | lrm: 1.00 | dt: 1192.60ms | tok/sec: 879,232 | bf16_mfu: 54.01 | total time: 18.73m | eta: 92.7m step 00945/05568 (16.97%) | loss: 2.780068 | lrm: 1.00 | dt: 1193.68ms | tok/sec: 878,436 | bf16_mfu: 53.96 | total time: 18.75m | eta: 92.7m step 00946/05568 (16.99%) | loss: 2.785021 | lrm: 1.00 | dt: 1193.74ms | tok/sec: 878,397 | bf16_mfu: 53.96 | total time: 18.77m | eta: 92.7m step 00947/05568 (17.01%) | loss: 2.796161 | lrm: 1.00 | dt: 1201.67ms | tok/sec: 872,602 | bf16_mfu: 53.60 | total time: 18.79m | eta: 92.7m step 00948/05568 (17.03%) | loss: 2.786699 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,306 | bf16_mfu: 53.95 | total time: 18.81m | eta: 92.6m step 00949/05568 (17.04%) | loss: 2.804329 | lrm: 1.00 | dt: 1193.78ms | tok/sec: 878,363 | bf16_mfu: 53.96 | total time: 18.83m | eta: 92.6m step 00950/05568 (17.06%) | loss: 2.808016 | lrm: 1.00 | dt: 1193.74ms | tok/sec: 878,398 | bf16_mfu: 53.96 | total time: 18.85m | eta: 92.6m step 00951/05568 (17.08%) | loss: 2.813359 | lrm: 1.00 | dt: 1194.54ms | tok/sec: 877,808 | bf16_mfu: 53.92 | total time: 18.87m | eta: 92.6m step 00952/05568 (17.10%) | loss: 2.814075 | lrm: 1.00 | dt: 1199.03ms | tok/sec: 874,517 | bf16_mfu: 53.72 | total time: 18.89m | eta: 92.6m step 00953/05568 (17.12%) | loss: 2.811456 | lrm: 1.00 | dt: 1196.93ms | tok/sec: 876,051 | bf16_mfu: 53.81 | total time: 18.91m | eta: 92.5m step 00954/05568 (17.13%) | loss: 2.817466 | lrm: 1.00 | dt: 1191.41ms | tok/sec: 880,110 | bf16_mfu: 54.06 | total time: 18.93m | eta: 92.5m step 00955/05568 (17.15%) | loss: 2.811630 | lrm: 1.00 | dt: 1197.38ms | tok/sec: 875,725 | bf16_mfu: 53.79 | total time: 18.95m | eta: 92.5m step 00956/05568 (17.17%) | loss: 2.805126 | lrm: 1.00 | dt: 1194.13ms | tok/sec: 878,107 | bf16_mfu: 53.94 | total time: 18.97m | eta: 92.5m step 00957/05568 (17.19%) | loss: 2.803382 | lrm: 1.00 | dt: 1194.91ms | tok/sec: 877,531 | bf16_mfu: 53.90 | total time: 18.99m | eta: 92.5m step 00958/05568 (17.21%) | loss: 2.799575 | lrm: 1.00 | dt: 1191.62ms | tok/sec: 879,961 | bf16_mfu: 54.05 | total time: 19.01m | eta: 92.4m step 00959/05568 (17.22%) | loss: 2.799524 | lrm: 1.00 | dt: 1193.41ms | tok/sec: 878,640 | bf16_mfu: 53.97 | total time: 19.03m | eta: 92.4m step 00960/05568 (17.24%) | loss: 2.803582 | lrm: 1.00 | dt: 1195.56ms | tok/sec: 877,056 | bf16_mfu: 53.88 | total time: 19.05m | eta: 92.4m step 00961/05568 (17.26%) | loss: 2.802690 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,635 | bf16_mfu: 53.91 | total time: 19.07m | eta: 92.4m step 00962/05568 (17.28%) | loss: 2.805701 | lrm: 1.00 | dt: 1194.06ms | tok/sec: 878,160 | bf16_mfu: 53.94 | total time: 19.09m | eta: 92.3m step 00963/05568 (17.30%) | loss: 2.813528 | lrm: 1.00 | dt: 1190.94ms | tok/sec: 880,457 | bf16_mfu: 54.08 | total time: 19.11m | eta: 92.3m step 00964/05568 (17.31%) | loss: 2.818396 | lrm: 1.00 | dt: 1195.86ms | tok/sec: 876,835 | bf16_mfu: 53.86 | total time: 19.13m | eta: 92.3m step 00965/05568 (17.33%) | loss: 2.810475 | lrm: 1.00 | dt: 1194.59ms | tok/sec: 877,771 | bf16_mfu: 53.92 | total time: 19.15m | eta: 92.3m step 00966/05568 (17.35%) | loss: 2.807668 | lrm: 1.00 | dt: 1194.49ms | tok/sec: 877,846 | bf16_mfu: 53.92 | total time: 19.17m | eta: 92.3m step 00967/05568 (17.37%) | loss: 2.807244 | lrm: 1.00 | dt: 1198.10ms | tok/sec: 875,196 | bf16_mfu: 53.76 | total time: 19.19m | eta: 92.2m step 00968/05568 (17.39%) | loss: 2.805217 | lrm: 1.00 | dt: 1191.15ms | tok/sec: 880,308 | bf16_mfu: 54.08 | total time: 19.21m | eta: 92.2m step 00969/05568 (17.40%) | loss: 2.811698 | lrm: 1.00 | dt: 1197.88ms | tok/sec: 875,357 | bf16_mfu: 53.77 | total time: 19.23m | eta: 92.2m step 00970/05568 (17.42%) | loss: 2.803536 | lrm: 1.00 | dt: 1193.68ms | tok/sec: 878,441 | bf16_mfu: 53.96 | total time: 19.25m | eta: 92.2m step 00971/05568 (17.44%) | loss: 2.787673 | lrm: 1.00 | dt: 1189.32ms | tok/sec: 881,657 | bf16_mfu: 54.16 | total time: 19.27m | eta: 92.2m step 00972/05568 (17.46%) | loss: 2.774614 | lrm: 1.00 | dt: 1195.71ms | tok/sec: 876,946 | bf16_mfu: 53.87 | total time: 19.29m | eta: 92.1m step 00973/05568 (17.47%) | loss: 2.777972 | lrm: 1.00 | dt: 1198.85ms | tok/sec: 874,649 | bf16_mfu: 53.73 | total time: 19.31m | eta: 92.1m step 00974/05568 (17.49%) | loss: 2.788556 | lrm: 1.00 | dt: 1197.11ms | tok/sec: 875,922 | bf16_mfu: 53.81 | total time: 19.33m | eta: 92.1m step 00975/05568 (17.51%) | loss: 2.788456 | lrm: 1.00 | dt: 1199.75ms | tok/sec: 873,993 | bf16_mfu: 53.69 | total time: 19.35m | eta: 92.1m step 00976/05568 (17.53%) | loss: 2.793878 | lrm: 1.00 | dt: 1195.90ms | tok/sec: 876,807 | bf16_mfu: 53.86 | total time: 19.37m | eta: 92.1m step 00977/05568 (17.55%) | loss: 2.801587 | lrm: 1.00 | dt: 1194.32ms | tok/sec: 877,970 | bf16_mfu: 53.93 | total time: 19.39m | eta: 92.0m step 00978/05568 (17.56%) | loss: 2.801501 | lrm: 1.00 | dt: 1194.60ms | tok/sec: 877,766 | bf16_mfu: 53.92 | total time: 19.41m | eta: 92.0m step 00979/05568 (17.58%) | loss: 2.806698 | lrm: 1.00 | dt: 1196.77ms | tok/sec: 876,173 | bf16_mfu: 53.82 | total time: 19.43m | eta: 92.0m step 00980/05568 (17.60%) | loss: 2.803645 | lrm: 1.00 | dt: 1199.25ms | tok/sec: 874,356 | bf16_mfu: 53.71 | total time: 19.45m | eta: 92.0m step 00981/05568 (17.62%) | loss: 2.797984 | lrm: 1.00 | dt: 1202.27ms | tok/sec: 872,162 | bf16_mfu: 53.57 | total time: 19.47m | eta: 92.0m step 00982/05568 (17.64%) | loss: 2.794278 | lrm: 1.00 | dt: 1196.67ms | tok/sec: 876,248 | bf16_mfu: 53.83 | total time: 19.49m | eta: 91.9m step 00983/05568 (17.65%) | loss: 2.784148 | lrm: 1.00 | dt: 1193.82ms | tok/sec: 878,339 | bf16_mfu: 53.95 | total time: 19.51m | eta: 91.9m step 00984/05568 (17.67%) | loss: 2.786709 | lrm: 1.00 | dt: 1189.30ms | tok/sec: 881,672 | bf16_mfu: 54.16 | total time: 19.53m | eta: 91.9m step 00985/05568 (17.69%) | loss: 2.779343 | lrm: 1.00 | dt: 1195.58ms | tok/sec: 877,045 | bf16_mfu: 53.87 | total time: 19.55m | eta: 91.9m step 00986/05568 (17.71%) | loss: 2.770686 | lrm: 1.00 | dt: 1191.61ms | tok/sec: 879,967 | bf16_mfu: 54.05 | total time: 19.57m | eta: 91.9m step 00987/05568 (17.73%) | loss: 2.770198 | lrm: 1.00 | dt: 1195.24ms | tok/sec: 877,291 | bf16_mfu: 53.89 | total time: 19.59m | eta: 91.8m step 00988/05568 (17.74%) | loss: 2.776110 | lrm: 1.00 | dt: 1190.45ms | tok/sec: 880,821 | bf16_mfu: 54.11 | total time: 19.61m | eta: 91.8m step 00989/05568 (17.76%) | loss: 2.781743 | lrm: 1.00 | dt: 1195.83ms | tok/sec: 876,861 | bf16_mfu: 53.86 | total time: 19.63m | eta: 91.8m step 00990/05568 (17.78%) | loss: 2.782491 | lrm: 1.00 | dt: 1192.25ms | tok/sec: 879,492 | bf16_mfu: 54.03 | total time: 19.65m | eta: 91.8m step 00991/05568 (17.80%) | loss: 2.776906 | lrm: 1.00 | dt: 1192.23ms | tok/sec: 879,506 | bf16_mfu: 54.03 | total time: 19.67m | eta: 91.7m step 00992/05568 (17.82%) | loss: 2.780022 | lrm: 1.00 | dt: 1190.13ms | tok/sec: 881,063 | bf16_mfu: 54.12 | total time: 19.69m | eta: 91.7m step 00993/05568 (17.83%) | loss: 2.783516 | lrm: 1.00 | dt: 1191.71ms | tok/sec: 879,892 | bf16_mfu: 54.05 | total time: 19.71m | eta: 91.7m step 00994/05568 (17.85%) | loss: 2.779617 | lrm: 1.00 | dt: 1192.54ms | tok/sec: 879,280 | bf16_mfu: 54.01 | total time: 19.73m | eta: 91.7m step 00995/05568 (17.87%) | loss: 2.769237 | lrm: 1.00 | dt: 1194.43ms | tok/sec: 877,889 | bf16_mfu: 53.93 | total time: 19.75m | eta: 91.7m step 00996/05568 (17.89%) | loss: 2.762411 | lrm: 1.00 | dt: 1193.90ms | tok/sec: 878,276 | bf16_mfu: 53.95 | total time: 19.77m | eta: 91.6m step 00997/05568 (17.91%) | loss: 2.756416 | lrm: 1.00 | dt: 1189.29ms | tok/sec: 881,685 | bf16_mfu: 54.16 | total time: 19.79m | eta: 91.6m step 00998/05568 (17.92%) | loss: 2.757201 | lrm: 1.00 | dt: 1200.28ms | tok/sec: 873,611 | bf16_mfu: 53.66 | total time: 19.81m | eta: 91.6m step 00999/05568 (17.94%) | loss: 2.762828 | lrm: 1.00 | dt: 1199.44ms | tok/sec: 874,218 | bf16_mfu: 53.70 | total time: 19.83m | eta: 91.6m step:1000/5568 val_bpb:0.846058 val_time:3.90s step 01000/05568 (17.96%) | loss: 2.767413 | lrm: 1.00 | dt: 1192.92ms | tok/sec: 879,000 | bf16_mfu: 53.99 | total time: 19.85m | eta: 91.6m step 01001/05568 (17.98%) | loss: 2.773915 | lrm: 1.00 | dt: 1195.40ms | tok/sec: 877,178 | bf16_mfu: 53.88 | total time: 19.87m | eta: 91.5m step 01002/05568 (18.00%) | loss: 2.790592 | lrm: 1.00 | dt: 1188.15ms | tok/sec: 882,529 | bf16_mfu: 54.21 | total time: 19.89m | eta: 91.5m step 01003/05568 (18.01%) | loss: 2.783499 | lrm: 1.00 | dt: 1200.56ms | tok/sec: 873,403 | bf16_mfu: 53.65 | total time: 19.91m | eta: 91.5m step 01004/05568 (18.03%) | loss: 2.771247 | lrm: 1.00 | dt: 1192.70ms | tok/sec: 879,160 | bf16_mfu: 54.00 | total time: 19.93m | eta: 91.5m step 01005/05568 (18.05%) | loss: 2.785055 | lrm: 1.00 | dt: 1193.84ms | tok/sec: 878,324 | bf16_mfu: 53.95 | total time: 19.95m | eta: 91.5m step 01006/05568 (18.07%) | loss: 2.790004 | lrm: 1.00 | dt: 1191.16ms | tok/sec: 880,300 | bf16_mfu: 54.07 | total time: 19.97m | eta: 91.4m step 01007/05568 (18.09%) | loss: 2.792470 | lrm: 1.00 | dt: 1204.77ms | tok/sec: 870,353 | bf16_mfu: 53.46 | total time: 19.99m | eta: 91.4m step 01008/05568 (18.10%) | loss: 2.792350 | lrm: 1.00 | dt: 1195.94ms | tok/sec: 876,777 | bf16_mfu: 53.86 | total time: 20.01m | eta: 91.4m step 01009/05568 (18.12%) | loss: 2.789223 | lrm: 1.00 | dt: 1200.79ms | tok/sec: 873,238 | bf16_mfu: 53.64 | total time: 20.03m | eta: 91.4m step 01010/05568 (18.14%) | loss: 2.793629 | lrm: 1.00 | dt: 1200.01ms | tok/sec: 873,803 | bf16_mfu: 53.68 | total time: 20.05m | eta: 91.4m step 01011/05568 (18.16%) | loss: 2.786947 | lrm: 1.00 | dt: 1194.83ms | tok/sec: 877,597 | bf16_mfu: 53.91 | total time: 20.07m | eta: 91.3m step 01012/05568 (18.18%) | loss: 2.801882 | lrm: 1.00 | dt: 1195.09ms | tok/sec: 877,403 | bf16_mfu: 53.90 | total time: 20.09m | eta: 91.3m step 01013/05568 (18.19%) | loss: 2.802323 | lrm: 1.00 | dt: 1192.27ms | tok/sec: 879,478 | bf16_mfu: 54.02 | total time: 20.11m | eta: 91.3m step 01014/05568 (18.21%) | loss: 2.808902 | lrm: 1.00 | dt: 1192.57ms | tok/sec: 879,255 | bf16_mfu: 54.01 | total time: 20.13m | eta: 91.3m step 01015/05568 (18.23%) | loss: 2.805756 | lrm: 1.00 | dt: 1194.30ms | tok/sec: 877,986 | bf16_mfu: 53.93 | total time: 20.15m | eta: 91.3m step 01016/05568 (18.25%) | loss: 2.807193 | lrm: 1.00 | dt: 1197.59ms | tok/sec: 875,573 | bf16_mfu: 53.78 | total time: 20.17m | eta: 91.2m step 01017/05568 (18.27%) | loss: 2.792797 | lrm: 1.00 | dt: 1193.49ms | tok/sec: 878,579 | bf16_mfu: 53.97 | total time: 20.19m | eta: 91.2m step 01018/05568 (18.28%) | loss: 2.787832 | lrm: 1.00 | dt: 1192.68ms | tok/sec: 879,178 | bf16_mfu: 54.01 | total time: 20.21m | eta: 91.2m step 01019/05568 (18.30%) | loss: 2.784890 | lrm: 1.00 | dt: 1192.66ms | tok/sec: 879,194 | bf16_mfu: 54.01 | total time: 20.23m | eta: 91.2m step 01020/05568 (18.32%) | loss: 2.779863 | lrm: 1.00 | dt: 1191.76ms | tok/sec: 879,857 | bf16_mfu: 54.05 | total time: 20.25m | eta: 91.1m step 01021/05568 (18.34%) | loss: 2.785946 | lrm: 1.00 | dt: 1194.24ms | tok/sec: 878,025 | bf16_mfu: 53.94 | total time: 20.27m | eta: 91.1m step 01022/05568 (18.35%) | loss: 2.781734 | lrm: 1.00 | dt: 1194.73ms | tok/sec: 877,671 | bf16_mfu: 53.91 | total time: 20.29m | eta: 91.1m step 01023/05568 (18.37%) | loss: 2.782982 | lrm: 1.00 | dt: 1192.50ms | tok/sec: 879,308 | bf16_mfu: 54.01 | total time: 20.31m | eta: 91.1m step 01024/05568 (18.39%) | loss: 2.782674 | lrm: 1.00 | dt: 1192.86ms | tok/sec: 879,045 | bf16_mfu: 54.00 | total time: 20.33m | eta: 91.1m step 01025/05568 (18.41%) | loss: 2.782117 | lrm: 1.00 | dt: 1192.40ms | tok/sec: 879,385 | bf16_mfu: 54.02 | total time: 20.35m | eta: 91.0m step 01026/05568 (18.43%) | loss: 2.783049 | lrm: 1.00 | dt: 1195.60ms | tok/sec: 877,026 | bf16_mfu: 53.87 | total time: 20.37m | eta: 91.0m step 01027/05568 (18.44%) | loss: 2.792691 | lrm: 1.00 | dt: 1197.67ms | tok/sec: 875,512 | bf16_mfu: 53.78 | total time: 20.39m | eta: 91.0m step 01028/05568 (18.46%) | loss: 2.793141 | lrm: 1.00 | dt: 1193.17ms | tok/sec: 878,814 | bf16_mfu: 53.98 | total time: 20.41m | eta: 91.0m step 01029/05568 (18.48%) | loss: 2.794263 | lrm: 1.00 | dt: 1196.70ms | tok/sec: 876,220 | bf16_mfu: 53.82 | total time: 20.43m | eta: 91.0m step 01030/05568 (18.50%) | loss: 2.796611 | lrm: 1.00 | dt: 1194.01ms | tok/sec: 878,199 | bf16_mfu: 53.95 | total time: 20.45m | eta: 90.9m step 01031/05568 (18.52%) | loss: 2.803244 | lrm: 1.00 | dt: 1195.99ms | tok/sec: 876,745 | bf16_mfu: 53.86 | total time: 20.47m | eta: 90.9m step 01032/05568 (18.53%) | loss: 2.794952 | lrm: 1.00 | dt: 1199.32ms | tok/sec: 874,312 | bf16_mfu: 53.71 | total time: 20.49m | eta: 90.9m step 01033/05568 (18.55%) | loss: 2.789626 | lrm: 1.00 | dt: 1190.19ms | tok/sec: 881,015 | bf16_mfu: 54.12 | total time: 20.51m | eta: 90.9m step 01034/05568 (18.57%) | loss: 2.789690 | lrm: 1.00 | dt: 1196.43ms | tok/sec: 876,422 | bf16_mfu: 53.84 | total time: 20.53m | eta: 90.9m step 01035/05568 (18.59%) | loss: 2.777087 | lrm: 1.00 | dt: 1194.16ms | tok/sec: 878,085 | bf16_mfu: 53.94 | total time: 20.55m | eta: 90.8m step 01036/05568 (18.61%) | loss: 2.770101 | lrm: 1.00 | dt: 1190.40ms | tok/sec: 880,859 | bf16_mfu: 54.11 | total time: 20.56m | eta: 90.8m step 01037/05568 (18.62%) | loss: 2.767008 | lrm: 1.00 | dt: 1194.50ms | tok/sec: 877,835 | bf16_mfu: 53.92 | total time: 20.58m | eta: 90.8m step 01038/05568 (18.64%) | loss: 2.776395 | lrm: 1.00 | dt: 1194.53ms | tok/sec: 877,816 | bf16_mfu: 53.92 | total time: 20.60m | eta: 90.8m step 01039/05568 (18.66%) | loss: 2.784637 | lrm: 1.00 | dt: 1195.04ms | tok/sec: 877,442 | bf16_mfu: 53.90 | total time: 20.62m | eta: 90.8m step 01040/05568 (18.68%) | loss: 2.779231 | lrm: 1.00 | dt: 1196.83ms | tok/sec: 876,128 | bf16_mfu: 53.82 | total time: 20.64m | eta: 90.7m step 01041/05568 (18.70%) | loss: 2.787048 | lrm: 1.00 | dt: 1199.32ms | tok/sec: 874,307 | bf16_mfu: 53.71 | total time: 20.66m | eta: 90.7m step 01042/05568 (18.71%) | loss: 2.772489 | lrm: 1.00 | dt: 1193.71ms | tok/sec: 878,414 | bf16_mfu: 53.96 | total time: 20.68m | eta: 90.7m step 01043/05568 (18.73%) | loss: 2.767533 | lrm: 1.00 | dt: 1193.65ms | tok/sec: 878,465 | bf16_mfu: 53.96 | total time: 20.70m | eta: 90.7m step 01044/05568 (18.75%) | loss: 2.760858 | lrm: 1.00 | dt: 1192.15ms | tok/sec: 879,567 | bf16_mfu: 54.03 | total time: 20.72m | eta: 90.7m step 01045/05568 (18.77%) | loss: 2.762784 | lrm: 1.00 | dt: 1194.61ms | tok/sec: 877,754 | bf16_mfu: 53.92 | total time: 20.74m | eta: 90.6m step 01046/05568 (18.79%) | loss: 2.777932 | lrm: 1.00 | dt: 1193.31ms | tok/sec: 878,713 | bf16_mfu: 53.98 | total time: 20.76m | eta: 90.6m step 01047/05568 (18.80%) | loss: 2.779979 | lrm: 1.00 | dt: 1192.31ms | tok/sec: 879,448 | bf16_mfu: 54.02 | total time: 20.78m | eta: 90.6m step 01048/05568 (18.82%) | loss: 2.776375 | lrm: 1.00 | dt: 1191.78ms | tok/sec: 879,843 | bf16_mfu: 54.05 | total time: 20.80m | eta: 90.6m step 01049/05568 (18.84%) | loss: 2.774689 | lrm: 1.00 | dt: 1196.11ms | tok/sec: 876,656 | bf16_mfu: 53.85 | total time: 20.82m | eta: 90.5m step 01050/05568 (18.86%) | loss: 2.777242 | lrm: 1.00 | dt: 1194.71ms | tok/sec: 877,683 | bf16_mfu: 53.91 | total time: 20.84m | eta: 90.5m step 01051/05568 (18.88%) | loss: 2.780628 | lrm: 1.00 | dt: 1191.38ms | tok/sec: 880,138 | bf16_mfu: 54.06 | total time: 20.86m | eta: 90.5m step 01052/05568 (18.89%) | loss: 2.777987 | lrm: 1.00 | dt: 1189.82ms | tok/sec: 881,287 | bf16_mfu: 54.14 | total time: 20.88m | eta: 90.5m step 01053/05568 (18.91%) | loss: 2.771935 | lrm: 1.00 | dt: 1196.03ms | tok/sec: 876,716 | bf16_mfu: 53.85 | total time: 20.90m | eta: 90.5m step 01054/05568 (18.93%) | loss: 2.767851 | lrm: 1.00 | dt: 1190.93ms | tok/sec: 880,467 | bf16_mfu: 54.09 | total time: 20.92m | eta: 90.4m step 01055/05568 (18.95%) | loss: 2.774073 | lrm: 1.00 | dt: 1198.72ms | tok/sec: 874,748 | bf16_mfu: 53.73 | total time: 20.94m | eta: 90.4m step 01056/05568 (18.97%) | loss: 2.784961 | lrm: 1.00 | dt: 1192.16ms | tok/sec: 879,561 | bf16_mfu: 54.03 | total time: 20.96m | eta: 90.4m step 01057/05568 (18.98%) | loss: 2.784668 | lrm: 1.00 | dt: 1193.93ms | tok/sec: 878,257 | bf16_mfu: 53.95 | total time: 20.98m | eta: 90.4m step 01058/05568 (19.00%) | loss: 2.781670 | lrm: 1.00 | dt: 1196.76ms | tok/sec: 876,178 | bf16_mfu: 53.82 | total time: 21.00m | eta: 90.4m step 01059/05568 (19.02%) | loss: 2.778910 | lrm: 1.00 | dt: 1189.45ms | tok/sec: 881,566 | bf16_mfu: 54.15 | total time: 21.02m | eta: 90.3m step 01060/05568 (19.04%) | loss: 2.763315 | lrm: 1.00 | dt: 1196.72ms | tok/sec: 876,209 | bf16_mfu: 53.82 | total time: 21.04m | eta: 90.3m step 01061/05568 (19.06%) | loss: 2.766641 | lrm: 1.00 | dt: 1197.71ms | tok/sec: 875,484 | bf16_mfu: 53.78 | total time: 21.06m | eta: 90.3m step 01062/05568 (19.07%) | loss: 2.770345 | lrm: 1.00 | dt: 1195.88ms | tok/sec: 876,826 | bf16_mfu: 53.86 | total time: 21.08m | eta: 90.3m step 01063/05568 (19.09%) | loss: 2.769772 | lrm: 1.00 | dt: 1187.38ms | tok/sec: 883,103 | bf16_mfu: 54.25 | total time: 21.10m | eta: 90.3m step 01064/05568 (19.11%) | loss: 2.756574 | lrm: 1.00 | dt: 1193.12ms | tok/sec: 878,854 | bf16_mfu: 53.99 | total time: 21.12m | eta: 90.2m step 01065/05568 (19.13%) | loss: 2.757452 | lrm: 1.00 | dt: 1194.55ms | tok/sec: 877,803 | bf16_mfu: 53.92 | total time: 21.14m | eta: 90.2m step 01066/05568 (19.15%) | loss: 2.761735 | lrm: 1.00 | dt: 1191.19ms | tok/sec: 880,276 | bf16_mfu: 54.07 | total time: 21.16m | eta: 90.2m step 01067/05568 (19.16%) | loss: 2.772070 | lrm: 1.00 | dt: 1194.51ms | tok/sec: 877,830 | bf16_mfu: 53.92 | total time: 21.18m | eta: 90.2m step 01068/05568 (19.18%) | loss: 2.769047 | lrm: 1.00 | dt: 1193.66ms | tok/sec: 878,453 | bf16_mfu: 53.96 | total time: 21.20m | eta: 90.2m step 01069/05568 (19.20%) | loss: 2.768522 | lrm: 1.00 | dt: 1198.04ms | tok/sec: 875,245 | bf16_mfu: 53.76 | total time: 21.22m | eta: 90.1m step 01070/05568 (19.22%) | loss: 2.772646 | lrm: 1.00 | dt: 1198.88ms | tok/sec: 874,626 | bf16_mfu: 53.73 | total time: 21.24m | eta: 90.1m step 01071/05568 (19.23%) | loss: 2.775163 | lrm: 1.00 | dt: 1188.33ms | tok/sec: 882,394 | bf16_mfu: 54.20 | total time: 21.26m | eta: 90.1m step 01072/05568 (19.25%) | loss: 2.772804 | lrm: 1.00 | dt: 1199.58ms | tok/sec: 874,121 | bf16_mfu: 53.70 | total time: 21.28m | eta: 90.1m step 01073/05568 (19.27%) | loss: 2.773391 | lrm: 1.00 | dt: 1194.27ms | tok/sec: 878,008 | bf16_mfu: 53.93 | total time: 21.30m | eta: 90.1m step 01074/05568 (19.29%) | loss: 2.763716 | lrm: 1.00 | dt: 1193.22ms | tok/sec: 878,778 | bf16_mfu: 53.98 | total time: 21.32m | eta: 90.0m step 01075/05568 (19.31%) | loss: 2.767285 | lrm: 1.00 | dt: 1195.14ms | tok/sec: 877,367 | bf16_mfu: 53.89 | total time: 21.34m | eta: 90.0m step 01076/05568 (19.32%) | loss: 2.777954 | lrm: 1.00 | dt: 1195.53ms | tok/sec: 877,076 | bf16_mfu: 53.88 | total time: 21.36m | eta: 90.0m step 01077/05568 (19.34%) | loss: 2.782040 | lrm: 1.00 | dt: 1191.38ms | tok/sec: 880,136 | bf16_mfu: 54.06 | total time: 21.38m | eta: 90.0m step 01078/05568 (19.36%) | loss: 2.767413 | lrm: 1.00 | dt: 1194.88ms | tok/sec: 877,556 | bf16_mfu: 53.91 | total time: 21.40m | eta: 90.0m step 01079/05568 (19.38%) | loss: 2.759614 | lrm: 1.00 | dt: 1196.01ms | tok/sec: 876,727 | bf16_mfu: 53.86 | total time: 21.42m | eta: 89.9m step 01080/05568 (19.40%) | loss: 2.762708 | lrm: 1.00 | dt: 1194.59ms | tok/sec: 877,772 | bf16_mfu: 53.92 | total time: 21.44m | eta: 89.9m step 01081/05568 (19.41%) | loss: 2.755177 | lrm: 1.00 | dt: 1195.06ms | tok/sec: 877,424 | bf16_mfu: 53.90 | total time: 21.46m | eta: 89.9m step 01082/05568 (19.43%) | loss: 2.767816 | lrm: 1.00 | dt: 1198.30ms | tok/sec: 875,051 | bf16_mfu: 53.75 | total time: 21.48m | eta: 89.9m step 01083/05568 (19.45%) | loss: 2.774227 | lrm: 1.00 | dt: 1191.00ms | tok/sec: 880,419 | bf16_mfu: 54.08 | total time: 21.50m | eta: 89.8m step 01084/05568 (19.47%) | loss: 2.783548 | lrm: 1.00 | dt: 1195.01ms | tok/sec: 877,464 | bf16_mfu: 53.90 | total time: 21.52m | eta: 89.8m step 01085/05568 (19.49%) | loss: 2.783475 | lrm: 1.00 | dt: 1191.83ms | tok/sec: 879,800 | bf16_mfu: 54.04 | total time: 21.54m | eta: 89.8m step 01086/05568 (19.50%) | loss: 2.778935 | lrm: 1.00 | dt: 1192.25ms | tok/sec: 879,495 | bf16_mfu: 54.03 | total time: 21.56m | eta: 89.8m step 01087/05568 (19.52%) | loss: 2.779015 | lrm: 1.00 | dt: 1191.65ms | tok/sec: 879,935 | bf16_mfu: 54.05 | total time: 21.58m | eta: 89.8m step 01088/05568 (19.54%) | loss: 2.776571 | lrm: 1.00 | dt: 1193.06ms | tok/sec: 878,894 | bf16_mfu: 53.99 | total time: 21.60m | eta: 89.7m step 01089/05568 (19.56%) | loss: 2.783849 | lrm: 1.00 | dt: 1190.26ms | tok/sec: 880,964 | bf16_mfu: 54.12 | total time: 21.62m | eta: 89.7m step 01090/05568 (19.58%) | loss: 2.793221 | lrm: 1.00 | dt: 1195.77ms | tok/sec: 876,901 | bf16_mfu: 53.87 | total time: 21.64m | eta: 89.7m step 01091/05568 (19.59%) | loss: 2.793563 | lrm: 1.00 | dt: 1192.93ms | tok/sec: 878,990 | bf16_mfu: 53.99 | total time: 21.66m | eta: 89.7m step 01092/05568 (19.61%) | loss: 2.787748 | lrm: 1.00 | dt: 1192.98ms | tok/sec: 878,953 | bf16_mfu: 53.99 | total time: 21.68m | eta: 89.7m step 01093/05568 (19.63%) | loss: 2.787339 | lrm: 1.00 | dt: 1194.18ms | tok/sec: 878,074 | bf16_mfu: 53.94 | total time: 21.70m | eta: 89.6m step 01094/05568 (19.65%) | loss: 2.776772 | lrm: 1.00 | dt: 1192.00ms | tok/sec: 879,674 | bf16_mfu: 54.04 | total time: 21.72m | eta: 89.6m step 01095/05568 (19.67%) | loss: 2.777792 | lrm: 1.00 | dt: 1197.12ms | tok/sec: 875,913 | bf16_mfu: 53.81 | total time: 21.74m | eta: 89.6m step 01096/05568 (19.68%) | loss: 2.771026 | lrm: 1.00 | dt: 1199.09ms | tok/sec: 874,473 | bf16_mfu: 53.72 | total time: 21.76m | eta: 89.6m step 01097/05568 (19.70%) | loss: 2.780690 | lrm: 1.00 | dt: 1194.81ms | tok/sec: 877,609 | bf16_mfu: 53.91 | total time: 21.78m | eta: 89.6m step 01098/05568 (19.72%) | loss: 2.772815 | lrm: 1.00 | dt: 1187.09ms | tok/sec: 883,315 | bf16_mfu: 54.26 | total time: 21.80m | eta: 89.5m step 01099/05568 (19.74%) | loss: 2.764717 | lrm: 1.00 | dt: 1193.69ms | tok/sec: 878,435 | bf16_mfu: 53.96 | total time: 21.82m | eta: 89.5m step 01100/05568 (19.76%) | loss: 2.764851 | lrm: 1.00 | dt: 1192.81ms | tok/sec: 879,079 | bf16_mfu: 54.00 | total time: 21.84m | eta: 89.5m step 01101/05568 (19.77%) | loss: 2.761405 | lrm: 1.00 | dt: 1191.56ms | tok/sec: 879,999 | bf16_mfu: 54.06 | total time: 21.86m | eta: 89.5m step 01102/05568 (19.79%) | loss: 2.771300 | lrm: 1.00 | dt: 1194.05ms | tok/sec: 878,169 | bf16_mfu: 53.94 | total time: 21.88m | eta: 89.5m step 01103/05568 (19.81%) | loss: 2.773465 | lrm: 1.00 | dt: 1196.79ms | tok/sec: 876,158 | bf16_mfu: 53.82 | total time: 21.90m | eta: 89.4m step 01104/05568 (19.83%) | loss: 2.766664 | lrm: 1.00 | dt: 1199.20ms | tok/sec: 874,396 | bf16_mfu: 53.71 | total time: 21.92m | eta: 89.4m step 01105/05568 (19.85%) | loss: 2.761403 | lrm: 1.00 | dt: 1199.51ms | tok/sec: 874,167 | bf16_mfu: 53.70 | total time: 21.94m | eta: 89.4m step 01106/05568 (19.86%) | loss: 2.769864 | lrm: 1.00 | dt: 1192.00ms | tok/sec: 879,679 | bf16_mfu: 54.04 | total time: 21.96m | eta: 89.4m step 01107/05568 (19.88%) | loss: 2.760402 | lrm: 1.00 | dt: 1194.75ms | tok/sec: 877,650 | bf16_mfu: 53.91 | total time: 21.98m | eta: 89.4m step 01108/05568 (19.90%) | loss: 2.751556 | lrm: 1.00 | dt: 1198.82ms | tok/sec: 874,673 | bf16_mfu: 53.73 | total time: 22.00m | eta: 89.3m step 01109/05568 (19.92%) | loss: 2.757173 | lrm: 1.00 | dt: 1195.78ms | tok/sec: 876,893 | bf16_mfu: 53.87 | total time: 22.02m | eta: 89.3m step 01110/05568 (19.94%) | loss: 2.750888 | lrm: 1.00 | dt: 1190.69ms | tok/sec: 880,645 | bf16_mfu: 54.10 | total time: 22.04m | eta: 89.3m step 01111/05568 (19.95%) | loss: 2.753368 | lrm: 1.00 | dt: 1196.99ms | tok/sec: 876,011 | bf16_mfu: 53.81 | total time: 22.06m | eta: 89.3m step 01112/05568 (19.97%) | loss: 2.750477 | lrm: 1.00 | dt: 1193.93ms | tok/sec: 878,254 | bf16_mfu: 53.95 | total time: 22.08m | eta: 89.3m step 01113/05568 (19.99%) | loss: 2.748571 | lrm: 1.00 | dt: 1194.11ms | tok/sec: 878,126 | bf16_mfu: 53.94 | total time: 22.10m | eta: 89.2m step 01114/05568 (20.01%) | loss: 2.748578 | lrm: 1.00 | dt: 1192.63ms | tok/sec: 879,211 | bf16_mfu: 54.01 | total time: 22.12m | eta: 89.2m step 01115/05568 (20.03%) | loss: 2.753982 | lrm: 1.00 | dt: 1197.21ms | tok/sec: 875,851 | bf16_mfu: 53.80 | total time: 22.14m | eta: 89.2m step 01116/05568 (20.04%) | loss: 2.749615 | lrm: 1.00 | dt: 1194.46ms | tok/sec: 877,867 | bf16_mfu: 53.93 | total time: 22.16m | eta: 89.2m step 01117/05568 (20.06%) | loss: 2.747448 | lrm: 1.00 | dt: 1195.08ms | tok/sec: 877,410 | bf16_mfu: 53.90 | total time: 22.18m | eta: 89.1m step 01118/05568 (20.08%) | loss: 2.735484 | lrm: 1.00 | dt: 1193.35ms | tok/sec: 878,680 | bf16_mfu: 53.98 | total time: 22.20m | eta: 89.1m step 01119/05568 (20.10%) | loss: 2.748265 | lrm: 1.00 | dt: 1192.57ms | tok/sec: 879,256 | bf16_mfu: 54.01 | total time: 22.22m | eta: 89.1m step 01120/05568 (20.11%) | loss: 2.759851 | lrm: 1.00 | dt: 1197.26ms | tok/sec: 875,814 | bf16_mfu: 53.80 | total time: 22.24m | eta: 89.1m step 01121/05568 (20.13%) | loss: 2.745264 | lrm: 1.00 | dt: 1196.25ms | tok/sec: 876,553 | bf16_mfu: 53.84 | total time: 22.26m | eta: 89.1m step 01122/05568 (20.15%) | loss: 2.743686 | lrm: 1.00 | dt: 1196.71ms | tok/sec: 876,213 | bf16_mfu: 53.82 | total time: 22.28m | eta: 89.0m step 01123/05568 (20.17%) | loss: 2.755598 | lrm: 1.00 | dt: 1190.66ms | tok/sec: 880,666 | bf16_mfu: 54.10 | total time: 22.30m | eta: 89.0m step 01124/05568 (20.19%) | loss: 2.754300 | lrm: 1.00 | dt: 1194.03ms | tok/sec: 878,181 | bf16_mfu: 53.94 | total time: 22.32m | eta: 89.0m step 01125/05568 (20.20%) | loss: 2.745083 | lrm: 1.00 | dt: 1193.78ms | tok/sec: 878,364 | bf16_mfu: 53.96 | total time: 22.34m | eta: 89.0m step 01126/05568 (20.22%) | loss: 2.750587 | lrm: 1.00 | dt: 1197.48ms | tok/sec: 875,652 | bf16_mfu: 53.79 | total time: 22.36m | eta: 89.0m step 01127/05568 (20.24%) | loss: 2.751662 | lrm: 1.00 | dt: 1194.13ms | tok/sec: 878,110 | bf16_mfu: 53.94 | total time: 22.38m | eta: 88.9m step 01128/05568 (20.26%) | loss: 2.761022 | lrm: 1.00 | dt: 1196.81ms | tok/sec: 876,141 | bf16_mfu: 53.82 | total time: 22.40m | eta: 88.9m step 01129/05568 (20.28%) | loss: 2.772772 | lrm: 1.00 | dt: 1196.94ms | tok/sec: 876,048 | bf16_mfu: 53.81 | total time: 22.42m | eta: 88.9m step 01130/05568 (20.29%) | loss: 2.753998 | lrm: 1.00 | dt: 1194.65ms | tok/sec: 877,727 | bf16_mfu: 53.92 | total time: 22.44m | eta: 88.9m step 01131/05568 (20.31%) | loss: 2.760294 | lrm: 1.00 | dt: 1195.04ms | tok/sec: 877,441 | bf16_mfu: 53.90 | total time: 22.46m | eta: 88.9m step 01132/05568 (20.33%) | loss: 2.763352 | lrm: 1.00 | dt: 1196.57ms | tok/sec: 876,314 | bf16_mfu: 53.83 | total time: 22.48m | eta: 88.8m step 01133/05568 (20.35%) | loss: 2.766788 | lrm: 1.00 | dt: 1194.19ms | tok/sec: 878,068 | bf16_mfu: 53.94 | total time: 22.50m | eta: 88.8m step 01134/05568 (20.37%) | loss: 2.754787 | lrm: 1.00 | dt: 1189.85ms | tok/sec: 881,267 | bf16_mfu: 54.13 | total time: 22.52m | eta: 88.8m step 01135/05568 (20.38%) | loss: 2.756157 | lrm: 1.00 | dt: 1194.82ms | tok/sec: 877,600 | bf16_mfu: 53.91 | total time: 22.54m | eta: 88.8m step 01136/05568 (20.40%) | loss: 2.753245 | lrm: 1.00 | dt: 1192.06ms | tok/sec: 879,637 | bf16_mfu: 54.03 | total time: 22.56m | eta: 88.8m step 01137/05568 (20.42%) | loss: 2.747811 | lrm: 1.00 | dt: 1190.69ms | tok/sec: 880,647 | bf16_mfu: 54.10 | total time: 22.58m | eta: 88.7m step 01138/05568 (20.44%) | loss: 2.753554 | lrm: 1.00 | dt: 1196.94ms | tok/sec: 876,049 | bf16_mfu: 53.81 | total time: 22.60m | eta: 88.7m step 01139/05568 (20.46%) | loss: 2.758442 | lrm: 1.00 | dt: 1194.14ms | tok/sec: 878,102 | bf16_mfu: 53.94 | total time: 22.62m | eta: 88.7m step 01140/05568 (20.47%) | loss: 2.742443 | lrm: 1.00 | dt: 1193.95ms | tok/sec: 878,238 | bf16_mfu: 53.95 | total time: 22.64m | eta: 88.7m step 01141/05568 (20.49%) | loss: 2.749170 | lrm: 1.00 | dt: 1195.09ms | tok/sec: 877,402 | bf16_mfu: 53.90 | total time: 22.65m | eta: 88.7m step 01142/05568 (20.51%) | loss: 2.739720 | lrm: 1.00 | dt: 1192.97ms | tok/sec: 878,963 | bf16_mfu: 53.99 | total time: 22.67m | eta: 88.6m step 01143/05568 (20.53%) | loss: 2.741956 | lrm: 1.00 | dt: 1192.74ms | tok/sec: 879,131 | bf16_mfu: 54.00 | total time: 22.69m | eta: 88.6m step 01144/05568 (20.55%) | loss: 2.736836 | lrm: 1.00 | dt: 1200.41ms | tok/sec: 873,512 | bf16_mfu: 53.66 | total time: 22.71m | eta: 88.6m step 01145/05568 (20.56%) | loss: 2.736159 | lrm: 1.00 | dt: 1194.37ms | tok/sec: 877,932 | bf16_mfu: 53.93 | total time: 22.73m | eta: 88.6m step 01146/05568 (20.58%) | loss: 2.734157 | lrm: 1.00 | dt: 1190.68ms | tok/sec: 880,650 | bf16_mfu: 54.10 | total time: 22.75m | eta: 88.6m step 01147/05568 (20.60%) | loss: 2.739499 | lrm: 1.00 | dt: 1190.16ms | tok/sec: 881,035 | bf16_mfu: 54.12 | total time: 22.77m | eta: 88.5m step 01148/05568 (20.62%) | loss: 2.751457 | lrm: 1.00 | dt: 1195.33ms | tok/sec: 877,225 | bf16_mfu: 53.89 | total time: 22.79m | eta: 88.5m step 01149/05568 (20.64%) | loss: 2.739167 | lrm: 1.00 | dt: 1195.56ms | tok/sec: 877,057 | bf16_mfu: 53.88 | total time: 22.81m | eta: 88.5m step 01150/05568 (20.65%) | loss: 2.734728 | lrm: 1.00 | dt: 1192.11ms | tok/sec: 879,596 | bf16_mfu: 54.03 | total time: 22.83m | eta: 88.5m step 01151/05568 (20.67%) | loss: 2.730676 | lrm: 1.00 | dt: 1194.16ms | tok/sec: 878,089 | bf16_mfu: 53.94 | total time: 22.85m | eta: 88.5m step 01152/05568 (20.69%) | loss: 2.730510 | lrm: 1.00 | dt: 1196.88ms | tok/sec: 876,090 | bf16_mfu: 53.82 | total time: 22.87m | eta: 88.4m step 01153/05568 (20.71%) | loss: 2.726085 | lrm: 1.00 | dt: 1192.31ms | tok/sec: 879,447 | bf16_mfu: 54.02 | total time: 22.89m | eta: 88.4m step 01154/05568 (20.73%) | loss: 2.732625 | lrm: 1.00 | dt: 1192.48ms | tok/sec: 879,321 | bf16_mfu: 54.01 | total time: 22.91m | eta: 88.4m step 01155/05568 (20.74%) | loss: 2.738047 | lrm: 1.00 | dt: 1195.87ms | tok/sec: 876,828 | bf16_mfu: 53.86 | total time: 22.93m | eta: 88.4m step 01156/05568 (20.76%) | loss: 2.745684 | lrm: 1.00 | dt: 1190.51ms | tok/sec: 880,781 | bf16_mfu: 54.10 | total time: 22.95m | eta: 88.3m step 01157/05568 (20.78%) | loss: 2.739533 | lrm: 1.00 | dt: 1193.25ms | tok/sec: 878,755 | bf16_mfu: 53.98 | total time: 22.97m | eta: 88.3m step 01158/05568 (20.80%) | loss: 2.728708 | lrm: 1.00 | dt: 1194.24ms | tok/sec: 878,026 | bf16_mfu: 53.94 | total time: 22.99m | eta: 88.3m step 01159/05568 (20.82%) | loss: 2.730387 | lrm: 1.00 | dt: 1192.45ms | tok/sec: 879,343 | bf16_mfu: 54.02 | total time: 23.01m | eta: 88.3m step 01160/05568 (20.83%) | loss: 2.722288 | lrm: 1.00 | dt: 1197.82ms | tok/sec: 875,403 | bf16_mfu: 53.77 | total time: 23.03m | eta: 88.3m step 01161/05568 (20.85%) | loss: 2.733684 | lrm: 1.00 | dt: 1190.02ms | tok/sec: 881,138 | bf16_mfu: 54.13 | total time: 23.05m | eta: 88.2m step 01162/05568 (20.87%) | loss: 2.731920 | lrm: 1.00 | dt: 1190.27ms | tok/sec: 880,953 | bf16_mfu: 54.11 | total time: 23.07m | eta: 88.2m step 01163/05568 (20.89%) | loss: 2.727091 | lrm: 1.00 | dt: 1191.82ms | tok/sec: 879,813 | bf16_mfu: 54.04 | total time: 23.09m | eta: 88.2m step 01164/05568 (20.91%) | loss: 2.736528 | lrm: 1.00 | dt: 1202.49ms | tok/sec: 872,004 | bf16_mfu: 53.57 | total time: 23.11m | eta: 88.2m step 01165/05568 (20.92%) | loss: 2.738145 | lrm: 1.00 | dt: 1196.25ms | tok/sec: 876,553 | bf16_mfu: 53.84 | total time: 23.13m | eta: 88.2m step 01166/05568 (20.94%) | loss: 2.741955 | lrm: 1.00 | dt: 1196.95ms | tok/sec: 876,041 | bf16_mfu: 53.81 | total time: 23.15m | eta: 88.1m step 01167/05568 (20.96%) | loss: 2.730801 | lrm: 1.00 | dt: 1189.44ms | tok/sec: 881,567 | bf16_mfu: 54.15 | total time: 23.17m | eta: 88.1m step 01168/05568 (20.98%) | loss: 2.739742 | lrm: 1.00 | dt: 1195.45ms | tok/sec: 877,136 | bf16_mfu: 53.88 | total time: 23.19m | eta: 88.1m step 01169/05568 (20.99%) | loss: 2.739570 | lrm: 1.00 | dt: 1196.29ms | tok/sec: 876,525 | bf16_mfu: 53.84 | total time: 23.21m | eta: 88.1m step 01170/05568 (21.01%) | loss: 2.740930 | lrm: 1.00 | dt: 1195.93ms | tok/sec: 876,788 | bf16_mfu: 53.86 | total time: 23.23m | eta: 88.1m step 01171/05568 (21.03%) | loss: 2.737997 | lrm: 1.00 | dt: 1194.90ms | tok/sec: 877,539 | bf16_mfu: 53.91 | total time: 23.25m | eta: 88.0m step 01172/05568 (21.05%) | loss: 2.747834 | lrm: 1.00 | dt: 1192.66ms | tok/sec: 879,187 | bf16_mfu: 54.01 | total time: 23.27m | eta: 88.0m step 01173/05568 (21.07%) | loss: 2.751132 | lrm: 1.00 | dt: 1194.54ms | tok/sec: 877,805 | bf16_mfu: 53.92 | total time: 23.29m | eta: 88.0m step 01174/05568 (21.08%) | loss: 2.756000 | lrm: 1.00 | dt: 1190.77ms | tok/sec: 880,584 | bf16_mfu: 54.09 | total time: 23.31m | eta: 88.0m step 01175/05568 (21.10%) | loss: 2.755458 | lrm: 1.00 | dt: 1195.20ms | tok/sec: 877,325 | bf16_mfu: 53.89 | total time: 23.33m | eta: 88.0m step 01176/05568 (21.12%) | loss: 2.762580 | lrm: 1.00 | dt: 1193.59ms | tok/sec: 878,502 | bf16_mfu: 53.96 | total time: 23.35m | eta: 87.9m step 01177/05568 (21.14%) | loss: 2.767164 | lrm: 1.00 | dt: 1193.00ms | tok/sec: 878,940 | bf16_mfu: 53.99 | total time: 23.37m | eta: 87.9m step 01178/05568 (21.16%) | loss: 2.771004 | lrm: 1.00 | dt: 1191.71ms | tok/sec: 879,891 | bf16_mfu: 54.05 | total time: 23.39m | eta: 87.9m step 01179/05568 (21.17%) | loss: 2.772177 | lrm: 1.00 | dt: 1191.71ms | tok/sec: 879,892 | bf16_mfu: 54.05 | total time: 23.41m | eta: 87.9m step 01180/05568 (21.19%) | loss: 2.776793 | lrm: 1.00 | dt: 1194.16ms | tok/sec: 878,088 | bf16_mfu: 53.94 | total time: 23.43m | eta: 87.9m step 01181/05568 (21.21%) | loss: 2.778237 | lrm: 1.00 | dt: 1197.15ms | tok/sec: 875,892 | bf16_mfu: 53.80 | total time: 23.45m | eta: 87.8m step 01182/05568 (21.23%) | loss: 2.782583 | lrm: 1.00 | dt: 1197.49ms | tok/sec: 875,642 | bf16_mfu: 53.79 | total time: 23.47m | eta: 87.8m step 01183/05568 (21.25%) | loss: 2.778834 | lrm: 1.00 | dt: 1194.98ms | tok/sec: 877,484 | bf16_mfu: 53.90 | total time: 23.49m | eta: 87.8m step 01184/05568 (21.26%) | loss: 2.780851 | lrm: 1.00 | dt: 1192.15ms | tok/sec: 879,568 | bf16_mfu: 54.03 | total time: 23.51m | eta: 87.8m step 01185/05568 (21.28%) | loss: 2.770096 | lrm: 1.00 | dt: 1191.52ms | tok/sec: 880,028 | bf16_mfu: 54.06 | total time: 23.53m | eta: 87.8m step 01186/05568 (21.30%) | loss: 2.762593 | lrm: 1.00 | dt: 1196.03ms | tok/sec: 876,710 | bf16_mfu: 53.85 | total time: 23.55m | eta: 87.7m step 01187/05568 (21.32%) | loss: 2.751785 | lrm: 1.00 | dt: 1193.31ms | tok/sec: 878,713 | bf16_mfu: 53.98 | total time: 23.57m | eta: 87.7m step 01188/05568 (21.34%) | loss: 2.749731 | lrm: 1.00 | dt: 1198.99ms | tok/sec: 874,551 | bf16_mfu: 53.72 | total time: 23.59m | eta: 87.7m step 01189/05568 (21.35%) | loss: 2.755928 | lrm: 1.00 | dt: 1192.13ms | tok/sec: 879,583 | bf16_mfu: 54.03 | total time: 23.61m | eta: 87.7m step 01190/05568 (21.37%) | loss: 2.750855 | lrm: 1.00 | dt: 1190.76ms | tok/sec: 880,591 | bf16_mfu: 54.09 | total time: 23.63m | eta: 87.7m step 01191/05568 (21.39%) | loss: 2.754265 | lrm: 1.00 | dt: 1192.22ms | tok/sec: 879,517 | bf16_mfu: 54.03 | total time: 23.65m | eta: 87.6m step 01192/05568 (21.41%) | loss: 2.750075 | lrm: 1.00 | dt: 1192.77ms | tok/sec: 879,111 | bf16_mfu: 54.00 | total time: 23.67m | eta: 87.6m step 01193/05568 (21.43%) | loss: 2.745415 | lrm: 1.00 | dt: 1194.12ms | tok/sec: 878,116 | bf16_mfu: 53.94 | total time: 23.69m | eta: 87.6m step 01194/05568 (21.44%) | loss: 2.743023 | lrm: 1.00 | dt: 1190.23ms | tok/sec: 880,984 | bf16_mfu: 54.12 | total time: 23.71m | eta: 87.6m step 01195/05568 (21.46%) | loss: 2.746310 | lrm: 1.00 | dt: 1197.33ms | tok/sec: 875,763 | bf16_mfu: 53.80 | total time: 23.73m | eta: 87.5m step 01196/05568 (21.48%) | loss: 2.749946 | lrm: 1.00 | dt: 1196.03ms | tok/sec: 876,715 | bf16_mfu: 53.85 | total time: 23.75m | eta: 87.5m step 01197/05568 (21.50%) | loss: 2.753078 | lrm: 1.00 | dt: 1191.18ms | tok/sec: 880,283 | bf16_mfu: 54.07 | total time: 23.77m | eta: 87.5m step 01198/05568 (21.52%) | loss: 2.741736 | lrm: 1.00 | dt: 1190.70ms | tok/sec: 880,641 | bf16_mfu: 54.10 | total time: 23.79m | eta: 87.5m step 01199/05568 (21.53%) | loss: 2.732885 | lrm: 1.00 | dt: 1192.71ms | tok/sec: 879,151 | bf16_mfu: 54.00 | total time: 23.81m | eta: 87.5m step 01200/05568 (21.55%) | loss: 2.726861 | lrm: 1.00 | dt: 1194.99ms | tok/sec: 877,477 | bf16_mfu: 53.90 | total time: 23.83m | eta: 87.4m step 01201/05568 (21.57%) | loss: 2.733778 | lrm: 1.00 | dt: 1196.22ms | tok/sec: 876,577 | bf16_mfu: 53.85 | total time: 23.85m | eta: 87.4m step 01202/05568 (21.59%) | loss: 2.727116 | lrm: 1.00 | dt: 1193.00ms | tok/sec: 878,941 | bf16_mfu: 53.99 | total time: 23.87m | eta: 87.4m step 01203/05568 (21.61%) | loss: 2.728088 | lrm: 1.00 | dt: 1191.11ms | tok/sec: 880,335 | bf16_mfu: 54.08 | total time: 23.89m | eta: 87.4m step 01204/05568 (21.62%) | loss: 2.728097 | lrm: 1.00 | dt: 1195.32ms | tok/sec: 877,236 | bf16_mfu: 53.89 | total time: 23.91m | eta: 87.4m step 01205/05568 (21.64%) | loss: 2.729952 | lrm: 1.00 | dt: 1192.52ms | tok/sec: 879,294 | bf16_mfu: 54.01 | total time: 23.93m | eta: 87.3m step 01206/05568 (21.66%) | loss: 2.738732 | lrm: 1.00 | dt: 1186.55ms | tok/sec: 883,721 | bf16_mfu: 54.28 | total time: 23.95m | eta: 87.3m step 01207/05568 (21.68%) | loss: 2.731513 | lrm: 1.00 | dt: 1202.33ms | tok/sec: 872,121 | bf16_mfu: 53.57 | total time: 23.97m | eta: 87.3m step 01208/05568 (21.70%) | loss: 2.734105 | lrm: 1.00 | dt: 1195.54ms | tok/sec: 877,070 | bf16_mfu: 53.88 | total time: 23.99m | eta: 87.3m step 01209/05568 (21.71%) | loss: 2.740504 | lrm: 1.00 | dt: 1192.62ms | tok/sec: 879,216 | bf16_mfu: 54.01 | total time: 24.01m | eta: 87.3m step 01210/05568 (21.73%) | loss: 2.740423 | lrm: 1.00 | dt: 1194.68ms | tok/sec: 877,704 | bf16_mfu: 53.92 | total time: 24.03m | eta: 87.2m step 01211/05568 (21.75%) | loss: 2.734053 | lrm: 1.00 | dt: 1189.88ms | tok/sec: 881,242 | bf16_mfu: 54.13 | total time: 24.05m | eta: 87.2m step 01212/05568 (21.77%) | loss: 2.743413 | lrm: 1.00 | dt: 1196.87ms | tok/sec: 876,097 | bf16_mfu: 53.82 | total time: 24.07m | eta: 87.2m step 01213/05568 (21.79%) | loss: 2.739566 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,398 | bf16_mfu: 54.02 | total time: 24.09m | eta: 87.2m step 01214/05568 (21.80%) | loss: 2.729898 | lrm: 1.00 | dt: 1199.32ms | tok/sec: 874,308 | bf16_mfu: 53.71 | total time: 24.11m | eta: 87.2m step 01215/05568 (21.82%) | loss: 2.732754 | lrm: 1.00 | dt: 1190.94ms | tok/sec: 880,458 | bf16_mfu: 54.08 | total time: 24.13m | eta: 87.1m step 01216/05568 (21.84%) | loss: 2.748814 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,307 | bf16_mfu: 53.95 | total time: 24.15m | eta: 87.1m step 01217/05568 (21.86%) | loss: 2.741353 | lrm: 1.00 | dt: 1195.15ms | tok/sec: 877,357 | bf16_mfu: 53.89 | total time: 24.17m | eta: 87.1m step 01218/05568 (21.88%) | loss: 2.732665 | lrm: 1.00 | dt: 1191.03ms | tok/sec: 880,390 | bf16_mfu: 54.08 | total time: 24.19m | eta: 87.1m step 01219/05568 (21.89%) | loss: 2.732708 | lrm: 1.00 | dt: 1196.29ms | tok/sec: 876,519 | bf16_mfu: 53.84 | total time: 24.21m | eta: 87.1m step 01220/05568 (21.91%) | loss: 2.728253 | lrm: 1.00 | dt: 1197.33ms | tok/sec: 875,758 | bf16_mfu: 53.80 | total time: 24.23m | eta: 87.0m step 01221/05568 (21.93%) | loss: 2.736132 | lrm: 1.00 | dt: 1191.49ms | tok/sec: 880,056 | bf16_mfu: 54.06 | total time: 24.25m | eta: 87.0m step 01222/05568 (21.95%) | loss: 2.741832 | lrm: 1.00 | dt: 1195.18ms | tok/sec: 877,338 | bf16_mfu: 53.89 | total time: 24.27m | eta: 87.0m step 01223/05568 (21.96%) | loss: 2.725246 | lrm: 1.00 | dt: 1192.35ms | tok/sec: 879,417 | bf16_mfu: 54.02 | total time: 24.29m | eta: 87.0m step 01224/05568 (21.98%) | loss: 2.723728 | lrm: 1.00 | dt: 1196.30ms | tok/sec: 876,516 | bf16_mfu: 53.84 | total time: 24.31m | eta: 87.0m step 01225/05568 (22.00%) | loss: 2.724105 | lrm: 1.00 | dt: 1189.31ms | tok/sec: 881,668 | bf16_mfu: 54.16 | total time: 24.33m | eta: 86.9m step 01226/05568 (22.02%) | loss: 2.728096 | lrm: 1.00 | dt: 1191.46ms | tok/sec: 880,076 | bf16_mfu: 54.06 | total time: 24.35m | eta: 86.9m step 01227/05568 (22.04%) | loss: 2.723387 | lrm: 1.00 | dt: 1192.39ms | tok/sec: 879,389 | bf16_mfu: 54.02 | total time: 24.37m | eta: 86.9m step 01228/05568 (22.05%) | loss: 2.723223 | lrm: 1.00 | dt: 1195.90ms | tok/sec: 876,806 | bf16_mfu: 53.86 | total time: 24.39m | eta: 86.9m step 01229/05568 (22.07%) | loss: 2.720493 | lrm: 1.00 | dt: 1193.24ms | tok/sec: 878,760 | bf16_mfu: 53.98 | total time: 24.41m | eta: 86.9m step 01230/05568 (22.09%) | loss: 2.724272 | lrm: 1.00 | dt: 1200.00ms | tok/sec: 873,813 | bf16_mfu: 53.68 | total time: 24.43m | eta: 86.8m step 01231/05568 (22.11%) | loss: 2.720980 | lrm: 1.00 | dt: 1191.97ms | tok/sec: 879,701 | bf16_mfu: 54.04 | total time: 24.45m | eta: 86.8m step 01232/05568 (22.13%) | loss: 2.739644 | lrm: 1.00 | dt: 1188.20ms | tok/sec: 882,488 | bf16_mfu: 54.21 | total time: 24.47m | eta: 86.8m step 01233/05568 (22.14%) | loss: 2.732648 | lrm: 1.00 | dt: 1197.09ms | tok/sec: 875,936 | bf16_mfu: 53.81 | total time: 24.49m | eta: 86.8m step 01234/05568 (22.16%) | loss: 2.739158 | lrm: 1.00 | dt: 1190.34ms | tok/sec: 880,902 | bf16_mfu: 54.11 | total time: 24.51m | eta: 86.7m step 01235/05568 (22.18%) | loss: 2.755059 | lrm: 1.00 | dt: 1196.18ms | tok/sec: 876,603 | bf16_mfu: 53.85 | total time: 24.53m | eta: 86.7m step 01236/05568 (22.20%) | loss: 2.758988 | lrm: 1.00 | dt: 1189.07ms | tok/sec: 881,841 | bf16_mfu: 54.17 | total time: 24.55m | eta: 86.7m step 01237/05568 (22.22%) | loss: 2.763229 | lrm: 1.00 | dt: 1192.78ms | tok/sec: 879,100 | bf16_mfu: 54.00 | total time: 24.57m | eta: 86.7m step 01238/05568 (22.23%) | loss: 2.757814 | lrm: 1.00 | dt: 1187.88ms | tok/sec: 882,727 | bf16_mfu: 54.22 | total time: 24.58m | eta: 86.7m step 01239/05568 (22.25%) | loss: 2.750365 | lrm: 1.00 | dt: 1192.88ms | tok/sec: 879,027 | bf16_mfu: 54.00 | total time: 24.60m | eta: 86.6m step 01240/05568 (22.27%) | loss: 2.750976 | lrm: 1.00 | dt: 1190.30ms | tok/sec: 880,931 | bf16_mfu: 54.11 | total time: 24.62m | eta: 86.6m step 01241/05568 (22.29%) | loss: 2.751112 | lrm: 1.00 | dt: 1189.75ms | tok/sec: 881,340 | bf16_mfu: 54.14 | total time: 24.64m | eta: 86.6m step 01242/05568 (22.31%) | loss: 2.748160 | lrm: 1.00 | dt: 1195.88ms | tok/sec: 876,825 | bf16_mfu: 53.86 | total time: 24.66m | eta: 86.6m step 01243/05568 (22.32%) | loss: 2.759378 | lrm: 1.00 | dt: 1191.92ms | tok/sec: 879,736 | bf16_mfu: 54.04 | total time: 24.68m | eta: 86.6m step 01244/05568 (22.34%) | loss: 2.751406 | lrm: 1.00 | dt: 1189.60ms | tok/sec: 881,454 | bf16_mfu: 54.15 | total time: 24.70m | eta: 86.5m step 01245/05568 (22.36%) | loss: 2.759722 | lrm: 1.00 | dt: 1193.29ms | tok/sec: 878,727 | bf16_mfu: 53.98 | total time: 24.72m | eta: 86.5m step 01246/05568 (22.38%) | loss: 2.761919 | lrm: 1.00 | dt: 1191.32ms | tok/sec: 880,182 | bf16_mfu: 54.07 | total time: 24.74m | eta: 86.5m step 01247/05568 (22.40%) | loss: 2.772596 | lrm: 1.00 | dt: 1196.33ms | tok/sec: 876,493 | bf16_mfu: 53.84 | total time: 24.76m | eta: 86.5m step 01248/05568 (22.41%) | loss: 2.774206 | lrm: 1.00 | dt: 1200.20ms | tok/sec: 873,670 | bf16_mfu: 53.67 | total time: 24.78m | eta: 86.5m step 01249/05568 (22.43%) | loss: 2.773290 | lrm: 1.00 | dt: 1197.48ms | tok/sec: 875,654 | bf16_mfu: 53.79 | total time: 24.80m | eta: 86.4m step:1250/5568 val_bpb:0.832731 val_time:3.89s step 01250/05568 (22.45%) | loss: 2.763047 | lrm: 1.00 | dt: 1194.24ms | tok/sec: 878,025 | bf16_mfu: 53.94 | total time: 24.82m | eta: 86.4m step 01251/05568 (22.47%) | loss: 2.770827 | lrm: 1.00 | dt: 1193.21ms | tok/sec: 878,783 | bf16_mfu: 53.98 | total time: 24.84m | eta: 86.4m step 01252/05568 (22.49%) | loss: 2.768978 | lrm: 1.00 | dt: 1195.00ms | tok/sec: 877,472 | bf16_mfu: 53.90 | total time: 24.86m | eta: 86.4m step 01253/05568 (22.50%) | loss: 2.766813 | lrm: 1.00 | dt: 1188.28ms | tok/sec: 882,434 | bf16_mfu: 54.21 | total time: 24.88m | eta: 86.4m step 01254/05568 (22.52%) | loss: 2.766806 | lrm: 1.00 | dt: 1192.22ms | tok/sec: 879,512 | bf16_mfu: 54.03 | total time: 24.90m | eta: 86.3m step 01255/05568 (22.54%) | loss: 2.758100 | lrm: 1.00 | dt: 1194.07ms | tok/sec: 878,149 | bf16_mfu: 53.94 | total time: 24.92m | eta: 86.3m step 01256/05568 (22.56%) | loss: 2.754099 | lrm: 1.00 | dt: 1194.43ms | tok/sec: 877,887 | bf16_mfu: 53.93 | total time: 24.94m | eta: 86.3m step 01257/05568 (22.58%) | loss: 2.752494 | lrm: 1.00 | dt: 1195.81ms | tok/sec: 876,876 | bf16_mfu: 53.86 | total time: 24.96m | eta: 86.3m step 01258/05568 (22.59%) | loss: 2.759328 | lrm: 1.00 | dt: 1196.63ms | tok/sec: 876,273 | bf16_mfu: 53.83 | total time: 24.98m | eta: 86.3m step 01259/05568 (22.61%) | loss: 2.758254 | lrm: 1.00 | dt: 1196.96ms | tok/sec: 876,035 | bf16_mfu: 53.81 | total time: 25.00m | eta: 86.2m step 01260/05568 (22.63%) | loss: 2.752140 | lrm: 1.00 | dt: 1193.45ms | tok/sec: 878,606 | bf16_mfu: 53.97 | total time: 25.02m | eta: 86.2m step 01261/05568 (22.65%) | loss: 2.747606 | lrm: 1.00 | dt: 1193.81ms | tok/sec: 878,346 | bf16_mfu: 53.95 | total time: 25.04m | eta: 86.2m step 01262/05568 (22.67%) | loss: 2.739314 | lrm: 1.00 | dt: 1190.42ms | tok/sec: 880,844 | bf16_mfu: 54.11 | total time: 25.06m | eta: 86.2m step 01263/05568 (22.68%) | loss: 2.736990 | lrm: 1.00 | dt: 1192.63ms | tok/sec: 879,213 | bf16_mfu: 54.01 | total time: 25.08m | eta: 86.2m step 01264/05568 (22.70%) | loss: 2.732948 | lrm: 1.00 | dt: 1195.23ms | tok/sec: 877,301 | bf16_mfu: 53.89 | total time: 25.10m | eta: 86.1m step 01265/05568 (22.72%) | loss: 2.737269 | lrm: 1.00 | dt: 1192.29ms | tok/sec: 879,464 | bf16_mfu: 54.02 | total time: 25.12m | eta: 86.1m step 01266/05568 (22.74%) | loss: 2.746497 | lrm: 1.00 | dt: 1191.85ms | tok/sec: 879,787 | bf16_mfu: 54.04 | total time: 25.14m | eta: 86.1m step 01267/05568 (22.76%) | loss: 2.740495 | lrm: 1.00 | dt: 1193.13ms | tok/sec: 878,847 | bf16_mfu: 53.99 | total time: 25.16m | eta: 86.1m step 01268/05568 (22.77%) | loss: 2.734258 | lrm: 1.00 | dt: 1193.40ms | tok/sec: 878,642 | bf16_mfu: 53.97 | total time: 25.18m | eta: 86.1m step 01269/05568 (22.79%) | loss: 2.731393 | lrm: 1.00 | dt: 1192.32ms | tok/sec: 879,444 | bf16_mfu: 54.02 | total time: 25.20m | eta: 86.0m step 01270/05568 (22.81%) | loss: 2.733171 | lrm: 1.00 | dt: 1192.37ms | tok/sec: 879,406 | bf16_mfu: 54.02 | total time: 25.22m | eta: 86.0m step 01271/05568 (22.83%) | loss: 2.735819 | lrm: 1.00 | dt: 1192.52ms | tok/sec: 879,292 | bf16_mfu: 54.01 | total time: 25.24m | eta: 86.0m step 01272/05568 (22.84%) | loss: 2.732011 | lrm: 1.00 | dt: 1193.62ms | tok/sec: 878,484 | bf16_mfu: 53.96 | total time: 25.26m | eta: 86.0m step 01273/05568 (22.86%) | loss: 2.738898 | lrm: 1.00 | dt: 1195.77ms | tok/sec: 876,906 | bf16_mfu: 53.87 | total time: 25.28m | eta: 86.0m step 01274/05568 (22.88%) | loss: 2.730847 | lrm: 1.00 | dt: 1196.07ms | tok/sec: 876,686 | bf16_mfu: 53.85 | total time: 25.30m | eta: 85.9m step 01275/05568 (22.90%) | loss: 2.738403 | lrm: 1.00 | dt: 1190.64ms | tok/sec: 880,680 | bf16_mfu: 54.10 | total time: 25.32m | eta: 85.9m step 01276/05568 (22.92%) | loss: 2.741304 | lrm: 1.00 | dt: 1192.91ms | tok/sec: 879,008 | bf16_mfu: 54.00 | total time: 25.34m | eta: 85.9m step 01277/05568 (22.93%) | loss: 2.734033 | lrm: 1.00 | dt: 1195.42ms | tok/sec: 877,161 | bf16_mfu: 53.88 | total time: 25.36m | eta: 85.9m step 01278/05568 (22.95%) | loss: 2.739006 | lrm: 1.00 | dt: 1189.97ms | tok/sec: 881,180 | bf16_mfu: 54.13 | total time: 25.38m | eta: 85.8m step 01279/05568 (22.97%) | loss: 2.740815 | lrm: 1.00 | dt: 1191.54ms | tok/sec: 880,020 | bf16_mfu: 54.06 | total time: 25.40m | eta: 85.8m step 01280/05568 (22.99%) | loss: 2.736509 | lrm: 1.00 | dt: 1196.67ms | tok/sec: 876,247 | bf16_mfu: 53.83 | total time: 25.42m | eta: 85.8m step 01281/05568 (23.01%) | loss: 2.732609 | lrm: 1.00 | dt: 1191.58ms | tok/sec: 879,987 | bf16_mfu: 54.06 | total time: 25.44m | eta: 85.8m step 01282/05568 (23.02%) | loss: 2.739524 | lrm: 1.00 | dt: 1192.94ms | tok/sec: 878,984 | bf16_mfu: 53.99 | total time: 25.46m | eta: 85.8m step 01283/05568 (23.04%) | loss: 2.732687 | lrm: 1.00 | dt: 1199.61ms | tok/sec: 874,098 | bf16_mfu: 53.69 | total time: 25.48m | eta: 85.7m step 01284/05568 (23.06%) | loss: 2.723601 | lrm: 1.00 | dt: 1193.42ms | tok/sec: 878,629 | bf16_mfu: 53.97 | total time: 25.50m | eta: 85.7m step 01285/05568 (23.08%) | loss: 2.726010 | lrm: 1.00 | dt: 1194.05ms | tok/sec: 878,169 | bf16_mfu: 53.94 | total time: 25.52m | eta: 85.7m step 01286/05568 (23.10%) | loss: 2.726888 | lrm: 1.00 | dt: 1194.19ms | tok/sec: 878,062 | bf16_mfu: 53.94 | total time: 25.54m | eta: 85.7m step 01287/05568 (23.11%) | loss: 2.739754 | lrm: 1.00 | dt: 1193.71ms | tok/sec: 878,418 | bf16_mfu: 53.96 | total time: 25.56m | eta: 85.7m step 01288/05568 (23.13%) | loss: 2.739503 | lrm: 1.00 | dt: 1198.08ms | tok/sec: 875,214 | bf16_mfu: 53.76 | total time: 25.58m | eta: 85.6m step 01289/05568 (23.15%) | loss: 2.746950 | lrm: 1.00 | dt: 1192.67ms | tok/sec: 879,185 | bf16_mfu: 54.01 | total time: 25.60m | eta: 85.6m step 01290/05568 (23.17%) | loss: 2.750646 | lrm: 1.00 | dt: 1195.78ms | tok/sec: 876,894 | bf16_mfu: 53.87 | total time: 25.62m | eta: 85.6m step 01291/05568 (23.19%) | loss: 2.762283 | lrm: 1.00 | dt: 1195.36ms | tok/sec: 877,206 | bf16_mfu: 53.88 | total time: 25.64m | eta: 85.6m step 01292/05568 (23.20%) | loss: 2.757059 | lrm: 1.00 | dt: 1189.74ms | tok/sec: 881,345 | bf16_mfu: 54.14 | total time: 25.66m | eta: 85.6m step 01293/05568 (23.22%) | loss: 2.757681 | lrm: 1.00 | dt: 1192.74ms | tok/sec: 879,129 | bf16_mfu: 54.00 | total time: 25.68m | eta: 85.5m step 01294/05568 (23.24%) | loss: 2.761406 | lrm: 1.00 | dt: 1189.43ms | tok/sec: 881,579 | bf16_mfu: 54.15 | total time: 25.70m | eta: 85.5m step 01295/05568 (23.26%) | loss: 2.742025 | lrm: 1.00 | dt: 1191.83ms | tok/sec: 879,802 | bf16_mfu: 54.04 | total time: 25.72m | eta: 85.5m step 01296/05568 (23.28%) | loss: 2.739317 | lrm: 1.00 | dt: 1192.79ms | tok/sec: 879,095 | bf16_mfu: 54.00 | total time: 25.74m | eta: 85.5m step 01297/05568 (23.29%) | loss: 2.740763 | lrm: 1.00 | dt: 1198.63ms | tok/sec: 874,809 | bf16_mfu: 53.74 | total time: 25.76m | eta: 85.5m step 01298/05568 (23.31%) | loss: 2.747372 | lrm: 1.00 | dt: 1190.42ms | tok/sec: 880,841 | bf16_mfu: 54.11 | total time: 25.78m | eta: 85.4m step 01299/05568 (23.33%) | loss: 2.748113 | lrm: 1.00 | dt: 1196.10ms | tok/sec: 876,662 | bf16_mfu: 53.85 | total time: 25.80m | eta: 85.4m step 01300/05568 (23.35%) | loss: 2.744245 | lrm: 1.00 | dt: 1197.30ms | tok/sec: 875,787 | bf16_mfu: 53.80 | total time: 25.82m | eta: 85.4m step 01301/05568 (23.37%) | loss: 2.738874 | lrm: 1.00 | dt: 1196.15ms | tok/sec: 876,627 | bf16_mfu: 53.85 | total time: 25.84m | eta: 85.4m step 01302/05568 (23.38%) | loss: 2.732905 | lrm: 1.00 | dt: 1192.11ms | tok/sec: 879,599 | bf16_mfu: 54.03 | total time: 25.86m | eta: 85.4m step 01303/05568 (23.40%) | loss: 2.732375 | lrm: 1.00 | dt: 1195.73ms | tok/sec: 876,931 | bf16_mfu: 53.87 | total time: 25.88m | eta: 85.3m step 01304/05568 (23.42%) | loss: 2.717621 | lrm: 1.00 | dt: 1195.02ms | tok/sec: 877,457 | bf16_mfu: 53.90 | total time: 25.90m | eta: 85.3m step 01305/05568 (23.44%) | loss: 2.732721 | lrm: 1.00 | dt: 1193.84ms | tok/sec: 878,318 | bf16_mfu: 53.95 | total time: 25.92m | eta: 85.3m step 01306/05568 (23.46%) | loss: 2.740439 | lrm: 1.00 | dt: 1192.66ms | tok/sec: 879,192 | bf16_mfu: 54.01 | total time: 25.94m | eta: 85.3m step 01307/05568 (23.47%) | loss: 2.749129 | lrm: 1.00 | dt: 1193.85ms | tok/sec: 878,317 | bf16_mfu: 53.95 | total time: 25.96m | eta: 85.3m step 01308/05568 (23.49%) | loss: 2.744275 | lrm: 1.00 | dt: 1188.27ms | tok/sec: 882,439 | bf16_mfu: 54.21 | total time: 25.98m | eta: 85.2m step 01309/05568 (23.51%) | loss: 2.730360 | lrm: 1.00 | dt: 1196.23ms | tok/sec: 876,564 | bf16_mfu: 53.85 | total time: 26.00m | eta: 85.2m step 01310/05568 (23.53%) | loss: 2.736228 | lrm: 1.00 | dt: 1193.98ms | tok/sec: 878,216 | bf16_mfu: 53.95 | total time: 26.02m | eta: 85.2m step 01311/05568 (23.55%) | loss: 2.744353 | lrm: 1.00 | dt: 1188.40ms | tok/sec: 882,339 | bf16_mfu: 54.20 | total time: 26.04m | eta: 85.2m step 01312/05568 (23.56%) | loss: 2.753049 | lrm: 1.00 | dt: 1196.04ms | tok/sec: 876,708 | bf16_mfu: 53.85 | total time: 26.06m | eta: 85.2m step 01313/05568 (23.58%) | loss: 2.747993 | lrm: 1.00 | dt: 1197.69ms | tok/sec: 875,497 | bf16_mfu: 53.78 | total time: 26.08m | eta: 85.1m step 01314/05568 (23.60%) | loss: 2.743287 | lrm: 1.00 | dt: 1188.83ms | tok/sec: 882,022 | bf16_mfu: 54.18 | total time: 26.10m | eta: 85.1m step 01315/05568 (23.62%) | loss: 2.747533 | lrm: 1.00 | dt: 1198.09ms | tok/sec: 875,208 | bf16_mfu: 53.76 | total time: 26.12m | eta: 85.1m step 01316/05568 (23.64%) | loss: 2.755307 | lrm: 1.00 | dt: 1193.91ms | tok/sec: 878,267 | bf16_mfu: 53.95 | total time: 26.14m | eta: 85.1m step 01317/05568 (23.65%) | loss: 2.743822 | lrm: 1.00 | dt: 1185.11ms | tok/sec: 884,792 | bf16_mfu: 54.35 | total time: 26.16m | eta: 85.1m step 01318/05568 (23.67%) | loss: 2.736560 | lrm: 1.00 | dt: 1194.22ms | tok/sec: 878,039 | bf16_mfu: 53.94 | total time: 26.18m | eta: 85.0m step 01319/05568 (23.69%) | loss: 2.737752 | lrm: 1.00 | dt: 1190.74ms | tok/sec: 880,605 | bf16_mfu: 54.09 | total time: 26.20m | eta: 85.0m step 01320/05568 (23.71%) | loss: 2.736092 | lrm: 1.00 | dt: 1193.26ms | tok/sec: 878,750 | bf16_mfu: 53.98 | total time: 26.22m | eta: 85.0m step 01321/05568 (23.72%) | loss: 2.735031 | lrm: 1.00 | dt: 1188.92ms | tok/sec: 881,958 | bf16_mfu: 54.18 | total time: 26.24m | eta: 85.0m step 01322/05568 (23.74%) | loss: 2.746462 | lrm: 1.00 | dt: 1188.81ms | tok/sec: 882,038 | bf16_mfu: 54.18 | total time: 26.26m | eta: 85.0m step 01323/05568 (23.76%) | loss: 2.741092 | lrm: 1.00 | dt: 1196.00ms | tok/sec: 876,738 | bf16_mfu: 53.86 | total time: 26.28m | eta: 84.9m step 01324/05568 (23.78%) | loss: 2.729531 | lrm: 1.00 | dt: 1192.28ms | tok/sec: 879,473 | bf16_mfu: 54.02 | total time: 26.30m | eta: 84.9m step 01325/05568 (23.80%) | loss: 2.733104 | lrm: 1.00 | dt: 1191.16ms | tok/sec: 880,296 | bf16_mfu: 54.07 | total time: 26.32m | eta: 84.9m step 01326/05568 (23.81%) | loss: 2.718797 | lrm: 1.00 | dt: 1197.99ms | tok/sec: 875,281 | bf16_mfu: 53.77 | total time: 26.34m | eta: 84.9m step 01327/05568 (23.83%) | loss: 2.731915 | lrm: 1.00 | dt: 1192.52ms | tok/sec: 879,297 | bf16_mfu: 54.01 | total time: 26.36m | eta: 84.8m step 01328/05568 (23.85%) | loss: 2.732052 | lrm: 1.00 | dt: 1190.51ms | tok/sec: 880,781 | bf16_mfu: 54.10 | total time: 26.37m | eta: 84.8m step 01329/05568 (23.87%) | loss: 2.728431 | lrm: 1.00 | dt: 1191.97ms | tok/sec: 879,700 | bf16_mfu: 54.04 | total time: 26.39m | eta: 84.8m step 01330/05568 (23.89%) | loss: 2.728655 | lrm: 1.00 | dt: 1197.99ms | tok/sec: 875,278 | bf16_mfu: 53.77 | total time: 26.41m | eta: 84.8m step 01331/05568 (23.90%) | loss: 2.733719 | lrm: 1.00 | dt: 1196.23ms | tok/sec: 876,568 | bf16_mfu: 53.85 | total time: 26.43m | eta: 84.8m step 01332/05568 (23.92%) | loss: 2.729652 | lrm: 1.00 | dt: 1193.55ms | tok/sec: 878,533 | bf16_mfu: 53.97 | total time: 26.45m | eta: 84.7m step 01333/05568 (23.94%) | loss: 2.733052 | lrm: 1.00 | dt: 1195.30ms | tok/sec: 877,246 | bf16_mfu: 53.89 | total time: 26.47m | eta: 84.7m step 01334/05568 (23.96%) | loss: 2.741243 | lrm: 1.00 | dt: 1192.23ms | tok/sec: 879,509 | bf16_mfu: 54.03 | total time: 26.49m | eta: 84.7m step 01335/05568 (23.98%) | loss: 2.736778 | lrm: 1.00 | dt: 1192.91ms | tok/sec: 879,004 | bf16_mfu: 54.00 | total time: 26.51m | eta: 84.7m step 01336/05568 (23.99%) | loss: 2.739805 | lrm: 1.00 | dt: 1192.08ms | tok/sec: 879,622 | bf16_mfu: 54.03 | total time: 26.53m | eta: 84.7m step 01337/05568 (24.01%) | loss: 2.744087 | lrm: 1.00 | dt: 1195.54ms | tok/sec: 877,071 | bf16_mfu: 53.88 | total time: 26.55m | eta: 84.6m step 01338/05568 (24.03%) | loss: 2.737489 | lrm: 1.00 | dt: 1194.21ms | tok/sec: 878,052 | bf16_mfu: 53.94 | total time: 26.57m | eta: 84.6m step 01339/05568 (24.05%) | loss: 2.737997 | lrm: 1.00 | dt: 1197.62ms | tok/sec: 875,546 | bf16_mfu: 53.78 | total time: 26.59m | eta: 84.6m step 01340/05568 (24.07%) | loss: 2.737945 | lrm: 1.00 | dt: 1198.52ms | tok/sec: 874,893 | bf16_mfu: 53.74 | total time: 26.61m | eta: 84.6m step 01341/05568 (24.08%) | loss: 2.741464 | lrm: 1.00 | dt: 1197.89ms | tok/sec: 875,353 | bf16_mfu: 53.77 | total time: 26.63m | eta: 84.6m step 01342/05568 (24.10%) | loss: 2.732201 | lrm: 1.00 | dt: 1195.59ms | tok/sec: 877,034 | bf16_mfu: 53.87 | total time: 26.65m | eta: 84.5m step 01343/05568 (24.12%) | loss: 2.732461 | lrm: 1.00 | dt: 1196.25ms | tok/sec: 876,555 | bf16_mfu: 53.84 | total time: 26.67m | eta: 84.5m step 01344/05568 (24.14%) | loss: 2.734600 | lrm: 1.00 | dt: 1193.46ms | tok/sec: 878,604 | bf16_mfu: 53.97 | total time: 26.69m | eta: 84.5m step 01345/05568 (24.16%) | loss: 2.734949 | lrm: 1.00 | dt: 1195.29ms | tok/sec: 877,256 | bf16_mfu: 53.89 | total time: 26.71m | eta: 84.5m step 01346/05568 (24.17%) | loss: 2.719091 | lrm: 1.00 | dt: 1194.26ms | tok/sec: 878,010 | bf16_mfu: 53.93 | total time: 26.73m | eta: 84.5m step 01347/05568 (24.19%) | loss: 2.726174 | lrm: 1.00 | dt: 1194.22ms | tok/sec: 878,045 | bf16_mfu: 53.94 | total time: 26.75m | eta: 84.4m step 01348/05568 (24.21%) | loss: 2.710975 | lrm: 1.00 | dt: 1186.83ms | tok/sec: 883,509 | bf16_mfu: 54.27 | total time: 26.77m | eta: 84.4m step 01349/05568 (24.23%) | loss: 2.708749 | lrm: 1.00 | dt: 1194.53ms | tok/sec: 877,817 | bf16_mfu: 53.92 | total time: 26.79m | eta: 84.4m step 01350/05568 (24.25%) | loss: 2.710514 | lrm: 1.00 | dt: 1195.14ms | tok/sec: 877,365 | bf16_mfu: 53.89 | total time: 26.81m | eta: 84.4m step 01351/05568 (24.26%) | loss: 2.712861 | lrm: 1.00 | dt: 1192.75ms | tok/sec: 879,124 | bf16_mfu: 54.00 | total time: 26.83m | eta: 84.4m step 01352/05568 (24.28%) | loss: 2.710990 | lrm: 1.00 | dt: 1195.49ms | tok/sec: 877,107 | bf16_mfu: 53.88 | total time: 26.85m | eta: 84.3m step 01353/05568 (24.30%) | loss: 2.718061 | lrm: 1.00 | dt: 1191.38ms | tok/sec: 880,135 | bf16_mfu: 54.06 | total time: 26.87m | eta: 84.3m step 01354/05568 (24.32%) | loss: 2.710460 | lrm: 1.00 | dt: 1192.10ms | tok/sec: 879,602 | bf16_mfu: 54.03 | total time: 26.89m | eta: 84.3m step 01355/05568 (24.34%) | loss: 2.705717 | lrm: 1.00 | dt: 1194.32ms | tok/sec: 877,969 | bf16_mfu: 53.93 | total time: 26.91m | eta: 84.3m step 01356/05568 (24.35%) | loss: 2.707779 | lrm: 1.00 | dt: 1191.66ms | tok/sec: 879,931 | bf16_mfu: 54.05 | total time: 26.93m | eta: 84.3m step 01357/05568 (24.37%) | loss: 2.714252 | lrm: 1.00 | dt: 1195.81ms | tok/sec: 876,877 | bf16_mfu: 53.86 | total time: 26.95m | eta: 84.2m step 01358/05568 (24.39%) | loss: 2.720903 | lrm: 1.00 | dt: 1193.58ms | tok/sec: 878,512 | bf16_mfu: 53.96 | total time: 26.97m | eta: 84.2m step 01359/05568 (24.41%) | loss: 2.724382 | lrm: 1.00 | dt: 1194.29ms | tok/sec: 877,990 | bf16_mfu: 53.93 | total time: 26.99m | eta: 84.2m step 01360/05568 (24.43%) | loss: 2.722282 | lrm: 1.00 | dt: 1189.96ms | tok/sec: 881,189 | bf16_mfu: 54.13 | total time: 27.01m | eta: 84.2m step 01361/05568 (24.44%) | loss: 2.722920 | lrm: 1.00 | dt: 1194.26ms | tok/sec: 878,010 | bf16_mfu: 53.93 | total time: 27.03m | eta: 84.2m step 01362/05568 (24.46%) | loss: 2.725273 | lrm: 1.00 | dt: 1192.83ms | tok/sec: 879,066 | bf16_mfu: 54.00 | total time: 27.05m | eta: 84.1m step 01363/05568 (24.48%) | loss: 2.728771 | lrm: 1.00 | dt: 1193.76ms | tok/sec: 878,381 | bf16_mfu: 53.96 | total time: 27.07m | eta: 84.1m step 01364/05568 (24.50%) | loss: 2.727472 | lrm: 1.00 | dt: 1195.09ms | tok/sec: 877,404 | bf16_mfu: 53.90 | total time: 27.09m | eta: 84.1m step 01365/05568 (24.52%) | loss: 2.719756 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,262 | bf16_mfu: 53.95 | total time: 27.11m | eta: 84.1m step 01366/05568 (24.53%) | loss: 2.726863 | lrm: 1.00 | dt: 1194.76ms | tok/sec: 877,643 | bf16_mfu: 53.91 | total time: 27.13m | eta: 84.1m step 01367/05568 (24.55%) | loss: 2.720808 | lrm: 1.00 | dt: 1189.69ms | tok/sec: 881,382 | bf16_mfu: 54.14 | total time: 27.15m | eta: 84.0m step 01368/05568 (24.57%) | loss: 2.731856 | lrm: 1.00 | dt: 1190.15ms | tok/sec: 881,045 | bf16_mfu: 54.12 | total time: 27.17m | eta: 84.0m step 01369/05568 (24.59%) | loss: 2.725491 | lrm: 1.00 | dt: 1193.34ms | tok/sec: 878,690 | bf16_mfu: 53.98 | total time: 27.19m | eta: 84.0m step 01370/05568 (24.60%) | loss: 2.717958 | lrm: 1.00 | dt: 1196.85ms | tok/sec: 876,113 | bf16_mfu: 53.82 | total time: 27.21m | eta: 84.0m step 01371/05568 (24.62%) | loss: 2.711082 | lrm: 1.00 | dt: 1192.06ms | tok/sec: 879,632 | bf16_mfu: 54.03 | total time: 27.23m | eta: 84.0m step 01372/05568 (24.64%) | loss: 2.718252 | lrm: 1.00 | dt: 1191.37ms | tok/sec: 880,141 | bf16_mfu: 54.07 | total time: 27.25m | eta: 83.9m step 01373/05568 (24.66%) | loss: 2.728887 | lrm: 1.00 | dt: 1189.86ms | tok/sec: 881,261 | bf16_mfu: 54.13 | total time: 27.27m | eta: 83.9m step 01374/05568 (24.68%) | loss: 2.719661 | lrm: 1.00 | dt: 1191.85ms | tok/sec: 879,787 | bf16_mfu: 54.04 | total time: 27.29m | eta: 83.9m step 01375/05568 (24.69%) | loss: 2.719837 | lrm: 1.00 | dt: 1195.85ms | tok/sec: 876,848 | bf16_mfu: 53.86 | total time: 27.31m | eta: 83.9m step 01376/05568 (24.71%) | loss: 2.712315 | lrm: 1.00 | dt: 1196.66ms | tok/sec: 876,252 | bf16_mfu: 53.83 | total time: 27.33m | eta: 83.9m step 01377/05568 (24.73%) | loss: 2.711807 | lrm: 1.00 | dt: 1191.36ms | tok/sec: 880,149 | bf16_mfu: 54.07 | total time: 27.35m | eta: 83.8m step 01378/05568 (24.75%) | loss: 2.725381 | lrm: 1.00 | dt: 1191.57ms | tok/sec: 879,995 | bf16_mfu: 54.06 | total time: 27.37m | eta: 83.8m step 01379/05568 (24.77%) | loss: 2.727739 | lrm: 1.00 | dt: 1190.52ms | tok/sec: 880,773 | bf16_mfu: 54.10 | total time: 27.39m | eta: 83.8m step 01380/05568 (24.78%) | loss: 2.743042 | lrm: 1.00 | dt: 1192.70ms | tok/sec: 879,163 | bf16_mfu: 54.00 | total time: 27.41m | eta: 83.8m step 01381/05568 (24.80%) | loss: 2.742934 | lrm: 1.00 | dt: 1192.44ms | tok/sec: 879,350 | bf16_mfu: 54.02 | total time: 27.43m | eta: 83.7m step 01382/05568 (24.82%) | loss: 2.749169 | lrm: 1.00 | dt: 1186.63ms | tok/sec: 883,662 | bf16_mfu: 54.28 | total time: 27.45m | eta: 83.7m step 01383/05568 (24.84%) | loss: 2.747428 | lrm: 1.00 | dt: 1200.03ms | tok/sec: 873,789 | bf16_mfu: 53.67 | total time: 27.47m | eta: 83.7m step 01384/05568 (24.86%) | loss: 2.725088 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,395 | bf16_mfu: 54.02 | total time: 27.49m | eta: 83.7m step 01385/05568 (24.87%) | loss: 2.726518 | lrm: 1.00 | dt: 1191.29ms | tok/sec: 880,200 | bf16_mfu: 54.07 | total time: 27.51m | eta: 83.7m step 01386/05568 (24.89%) | loss: 2.730636 | lrm: 1.00 | dt: 1195.92ms | tok/sec: 876,797 | bf16_mfu: 53.86 | total time: 27.53m | eta: 83.6m step 01387/05568 (24.91%) | loss: 2.730941 | lrm: 1.00 | dt: 1192.74ms | tok/sec: 879,129 | bf16_mfu: 54.00 | total time: 27.55m | eta: 83.6m step 01388/05568 (24.93%) | loss: 2.731326 | lrm: 1.00 | dt: 1194.14ms | tok/sec: 878,101 | bf16_mfu: 53.94 | total time: 27.57m | eta: 83.6m step 01389/05568 (24.95%) | loss: 2.738636 | lrm: 1.00 | dt: 1192.11ms | tok/sec: 879,598 | bf16_mfu: 54.03 | total time: 27.59m | eta: 83.6m step 01390/05568 (24.96%) | loss: 2.738595 | lrm: 1.00 | dt: 1192.60ms | tok/sec: 879,236 | bf16_mfu: 54.01 | total time: 27.61m | eta: 83.6m step 01391/05568 (24.98%) | loss: 2.733354 | lrm: 1.00 | dt: 1190.26ms | tok/sec: 880,965 | bf16_mfu: 54.12 | total time: 27.63m | eta: 83.5m step 01392/05568 (25.00%) | loss: 2.726950 | lrm: 1.00 | dt: 1186.32ms | tok/sec: 883,891 | bf16_mfu: 54.30 | total time: 27.65m | eta: 83.5m step 01393/05568 (25.02%) | loss: 2.724753 | lrm: 1.00 | dt: 1193.02ms | tok/sec: 878,923 | bf16_mfu: 53.99 | total time: 27.67m | eta: 83.5m step 01394/05568 (25.04%) | loss: 2.719421 | lrm: 1.00 | dt: 1190.91ms | tok/sec: 880,485 | bf16_mfu: 54.09 | total time: 27.69m | eta: 83.5m step 01395/05568 (25.05%) | loss: 2.714806 | lrm: 1.00 | dt: 1192.98ms | tok/sec: 878,953 | bf16_mfu: 53.99 | total time: 27.71m | eta: 83.5m step 01396/05568 (25.07%) | loss: 2.707567 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,396 | bf16_mfu: 54.02 | total time: 27.73m | eta: 83.4m step 01397/05568 (25.09%) | loss: 2.701109 | lrm: 1.00 | dt: 1197.95ms | tok/sec: 875,306 | bf16_mfu: 53.77 | total time: 27.75m | eta: 83.4m step 01398/05568 (25.11%) | loss: 2.709633 | lrm: 1.00 | dt: 1192.58ms | tok/sec: 879,248 | bf16_mfu: 54.01 | total time: 27.77m | eta: 83.4m step 01399/05568 (25.13%) | loss: 2.702399 | lrm: 1.00 | dt: 1190.83ms | tok/sec: 880,544 | bf16_mfu: 54.09 | total time: 27.79m | eta: 83.4m step 01400/05568 (25.14%) | loss: 2.699621 | lrm: 1.00 | dt: 1199.36ms | tok/sec: 874,281 | bf16_mfu: 53.71 | total time: 27.81m | eta: 83.4m step 01401/05568 (25.16%) | loss: 2.697261 | lrm: 1.00 | dt: 1196.17ms | tok/sec: 876,607 | bf16_mfu: 53.85 | total time: 27.83m | eta: 83.3m step 01402/05568 (25.18%) | loss: 2.705958 | lrm: 1.00 | dt: 1193.81ms | tok/sec: 878,342 | bf16_mfu: 53.95 | total time: 27.85m | eta: 83.3m step 01403/05568 (25.20%) | loss: 2.710337 | lrm: 1.00 | dt: 1189.46ms | tok/sec: 881,553 | bf16_mfu: 54.15 | total time: 27.87m | eta: 83.3m step 01404/05568 (25.22%) | loss: 2.711264 | lrm: 1.00 | dt: 1193.71ms | tok/sec: 878,421 | bf16_mfu: 53.96 | total time: 27.89m | eta: 83.3m step 01405/05568 (25.23%) | loss: 2.708640 | lrm: 1.00 | dt: 1196.89ms | tok/sec: 876,084 | bf16_mfu: 53.82 | total time: 27.91m | eta: 83.3m step 01406/05568 (25.25%) | loss: 2.703866 | lrm: 1.00 | dt: 1192.19ms | tok/sec: 879,535 | bf16_mfu: 54.03 | total time: 27.93m | eta: 83.2m step 01407/05568 (25.27%) | loss: 2.697480 | lrm: 1.00 | dt: 1194.13ms | tok/sec: 878,106 | bf16_mfu: 53.94 | total time: 27.95m | eta: 83.2m step 01408/05568 (25.29%) | loss: 2.699894 | lrm: 1.00 | dt: 1190.15ms | tok/sec: 881,045 | bf16_mfu: 54.12 | total time: 27.97m | eta: 83.2m step 01409/05568 (25.31%) | loss: 2.698313 | lrm: 1.00 | dt: 1190.63ms | tok/sec: 880,692 | bf16_mfu: 54.10 | total time: 27.99m | eta: 83.2m step 01410/05568 (25.32%) | loss: 2.699516 | lrm: 1.00 | dt: 1197.68ms | tok/sec: 875,505 | bf16_mfu: 53.78 | total time: 28.01m | eta: 83.2m step 01411/05568 (25.34%) | loss: 2.701038 | lrm: 1.00 | dt: 1197.07ms | tok/sec: 875,950 | bf16_mfu: 53.81 | total time: 28.03m | eta: 83.1m step 01412/05568 (25.36%) | loss: 2.701346 | lrm: 1.00 | dt: 1192.72ms | tok/sec: 879,145 | bf16_mfu: 54.00 | total time: 28.05m | eta: 83.1m step 01413/05568 (25.38%) | loss: 2.703095 | lrm: 1.00 | dt: 1192.60ms | tok/sec: 879,233 | bf16_mfu: 54.01 | total time: 28.07m | eta: 83.1m step 01414/05568 (25.40%) | loss: 2.707198 | lrm: 1.00 | dt: 1192.63ms | tok/sec: 879,212 | bf16_mfu: 54.01 | total time: 28.09m | eta: 83.1m step 01415/05568 (25.41%) | loss: 2.707499 | lrm: 1.00 | dt: 1196.82ms | tok/sec: 876,136 | bf16_mfu: 53.82 | total time: 28.11m | eta: 83.1m step 01416/05568 (25.43%) | loss: 2.710903 | lrm: 1.00 | dt: 1192.03ms | tok/sec: 879,654 | bf16_mfu: 54.04 | total time: 28.13m | eta: 83.0m step 01417/05568 (25.45%) | loss: 2.707116 | lrm: 1.00 | dt: 1197.07ms | tok/sec: 875,952 | bf16_mfu: 53.81 | total time: 28.15m | eta: 83.0m step 01418/05568 (25.47%) | loss: 2.721771 | lrm: 1.00 | dt: 1194.54ms | tok/sec: 877,804 | bf16_mfu: 53.92 | total time: 28.17m | eta: 83.0m step 01419/05568 (25.48%) | loss: 2.720656 | lrm: 1.00 | dt: 1191.66ms | tok/sec: 879,926 | bf16_mfu: 54.05 | total time: 28.19m | eta: 83.0m step 01420/05568 (25.50%) | loss: 2.721974 | lrm: 1.00 | dt: 1192.06ms | tok/sec: 879,635 | bf16_mfu: 54.03 | total time: 28.20m | eta: 83.0m step 01421/05568 (25.52%) | loss: 2.718648 | lrm: 1.00 | dt: 1193.93ms | tok/sec: 878,256 | bf16_mfu: 53.95 | total time: 28.22m | eta: 82.9m step 01422/05568 (25.54%) | loss: 2.722108 | lrm: 1.00 | dt: 1193.37ms | tok/sec: 878,664 | bf16_mfu: 53.97 | total time: 28.24m | eta: 82.9m step 01423/05568 (25.56%) | loss: 2.724787 | lrm: 1.00 | dt: 1191.79ms | tok/sec: 879,831 | bf16_mfu: 54.05 | total time: 28.26m | eta: 82.9m step 01424/05568 (25.57%) | loss: 2.722130 | lrm: 1.00 | dt: 1197.47ms | tok/sec: 875,658 | bf16_mfu: 53.79 | total time: 28.28m | eta: 82.9m step 01425/05568 (25.59%) | loss: 2.727993 | lrm: 1.00 | dt: 1192.91ms | tok/sec: 879,010 | bf16_mfu: 54.00 | total time: 28.30m | eta: 82.9m step 01426/05568 (25.61%) | loss: 2.729351 | lrm: 1.00 | dt: 1193.44ms | tok/sec: 878,615 | bf16_mfu: 53.97 | total time: 28.32m | eta: 82.8m step 01427/05568 (25.63%) | loss: 2.721901 | lrm: 1.00 | dt: 1195.20ms | tok/sec: 877,321 | bf16_mfu: 53.89 | total time: 28.34m | eta: 82.8m step 01428/05568 (25.65%) | loss: 2.723339 | lrm: 1.00 | dt: 1198.47ms | tok/sec: 874,929 | bf16_mfu: 53.74 | total time: 28.36m | eta: 82.8m step 01429/05568 (25.66%) | loss: 2.718749 | lrm: 1.00 | dt: 1190.80ms | tok/sec: 880,566 | bf16_mfu: 54.09 | total time: 28.38m | eta: 82.8m step 01430/05568 (25.68%) | loss: 2.714338 | lrm: 1.00 | dt: 1194.33ms | tok/sec: 877,959 | bf16_mfu: 53.93 | total time: 28.40m | eta: 82.8m step 01431/05568 (25.70%) | loss: 2.711703 | lrm: 1.00 | dt: 1191.75ms | tok/sec: 879,860 | bf16_mfu: 54.05 | total time: 28.42m | eta: 82.7m step 01432/05568 (25.72%) | loss: 2.707338 | lrm: 1.00 | dt: 1194.06ms | tok/sec: 878,158 | bf16_mfu: 53.94 | total time: 28.44m | eta: 82.7m step 01433/05568 (25.74%) | loss: 2.708035 | lrm: 1.00 | dt: 1186.81ms | tok/sec: 883,522 | bf16_mfu: 54.27 | total time: 28.46m | eta: 82.7m step 01434/05568 (25.75%) | loss: 2.715688 | lrm: 1.00 | dt: 1196.39ms | tok/sec: 876,452 | bf16_mfu: 53.84 | total time: 28.48m | eta: 82.7m step 01435/05568 (25.77%) | loss: 2.725204 | lrm: 1.00 | dt: 1187.13ms | tok/sec: 883,283 | bf16_mfu: 54.26 | total time: 28.50m | eta: 82.6m step 01436/05568 (25.79%) | loss: 2.727688 | lrm: 1.00 | dt: 1197.79ms | tok/sec: 875,422 | bf16_mfu: 53.78 | total time: 28.52m | eta: 82.6m step 01437/05568 (25.81%) | loss: 2.725811 | lrm: 1.00 | dt: 1193.84ms | tok/sec: 878,322 | bf16_mfu: 53.95 | total time: 28.54m | eta: 82.6m step 01438/05568 (25.83%) | loss: 2.722336 | lrm: 1.00 | dt: 1191.51ms | tok/sec: 880,041 | bf16_mfu: 54.06 | total time: 28.56m | eta: 82.6m step 01439/05568 (25.84%) | loss: 2.719537 | lrm: 1.00 | dt: 1194.91ms | tok/sec: 877,534 | bf16_mfu: 53.90 | total time: 28.58m | eta: 82.6m step 01440/05568 (25.86%) | loss: 2.713986 | lrm: 1.00 | dt: 1199.57ms | tok/sec: 874,127 | bf16_mfu: 53.70 | total time: 28.60m | eta: 82.5m step 01441/05568 (25.88%) | loss: 2.723107 | lrm: 1.00 | dt: 1192.21ms | tok/sec: 879,524 | bf16_mfu: 54.03 | total time: 28.62m | eta: 82.5m step 01442/05568 (25.90%) | loss: 2.721888 | lrm: 1.00 | dt: 1197.92ms | tok/sec: 875,332 | bf16_mfu: 53.77 | total time: 28.64m | eta: 82.5m step 01443/05568 (25.92%) | loss: 2.722042 | lrm: 1.00 | dt: 1193.31ms | tok/sec: 878,709 | bf16_mfu: 53.98 | total time: 28.66m | eta: 82.5m step 01444/05568 (25.93%) | loss: 2.726933 | lrm: 1.00 | dt: 1197.25ms | tok/sec: 875,819 | bf16_mfu: 53.80 | total time: 28.68m | eta: 82.5m step 01445/05568 (25.95%) | loss: 2.712360 | lrm: 1.00 | dt: 1192.66ms | tok/sec: 879,188 | bf16_mfu: 54.01 | total time: 28.70m | eta: 82.4m step 01446/05568 (25.97%) | loss: 2.711043 | lrm: 1.00 | dt: 1190.76ms | tok/sec: 880,591 | bf16_mfu: 54.09 | total time: 28.72m | eta: 82.4m step 01447/05568 (25.99%) | loss: 2.708726 | lrm: 1.00 | dt: 1196.64ms | tok/sec: 876,268 | bf16_mfu: 53.83 | total time: 28.74m | eta: 82.4m step 01448/05568 (26.01%) | loss: 2.695911 | lrm: 1.00 | dt: 1194.31ms | tok/sec: 877,976 | bf16_mfu: 53.93 | total time: 28.76m | eta: 82.4m step 01449/05568 (26.02%) | loss: 2.690671 | lrm: 1.00 | dt: 1193.23ms | tok/sec: 878,772 | bf16_mfu: 53.98 | total time: 28.78m | eta: 82.4m step 01450/05568 (26.04%) | loss: 2.688761 | lrm: 1.00 | dt: 1191.58ms | tok/sec: 879,985 | bf16_mfu: 54.06 | total time: 28.80m | eta: 82.3m step 01451/05568 (26.06%) | loss: 2.699263 | lrm: 1.00 | dt: 1195.42ms | tok/sec: 877,160 | bf16_mfu: 53.88 | total time: 28.82m | eta: 82.3m step 01452/05568 (26.08%) | loss: 2.696465 | lrm: 1.00 | dt: 1187.99ms | tok/sec: 882,649 | bf16_mfu: 54.22 | total time: 28.84m | eta: 82.3m step 01453/05568 (26.10%) | loss: 2.701311 | lrm: 1.00 | dt: 1199.81ms | tok/sec: 873,948 | bf16_mfu: 53.68 | total time: 28.86m | eta: 82.3m step 01454/05568 (26.11%) | loss: 2.711625 | lrm: 1.00 | dt: 1196.65ms | tok/sec: 876,260 | bf16_mfu: 53.83 | total time: 28.88m | eta: 82.3m step 01455/05568 (26.13%) | loss: 2.714089 | lrm: 1.00 | dt: 1194.88ms | tok/sec: 877,554 | bf16_mfu: 53.91 | total time: 28.90m | eta: 82.2m step 01456/05568 (26.15%) | loss: 2.723253 | lrm: 1.00 | dt: 1192.65ms | tok/sec: 879,196 | bf16_mfu: 54.01 | total time: 28.92m | eta: 82.2m step 01457/05568 (26.17%) | loss: 2.713560 | lrm: 1.00 | dt: 1191.74ms | tok/sec: 879,870 | bf16_mfu: 54.05 | total time: 28.94m | eta: 82.2m step 01458/05568 (26.19%) | loss: 2.721464 | lrm: 1.00 | dt: 1195.53ms | tok/sec: 877,077 | bf16_mfu: 53.88 | total time: 28.96m | eta: 82.2m step 01459/05568 (26.20%) | loss: 2.716350 | lrm: 1.00 | dt: 1190.20ms | tok/sec: 881,009 | bf16_mfu: 54.12 | total time: 28.98m | eta: 82.2m step 01460/05568 (26.22%) | loss: 2.716567 | lrm: 1.00 | dt: 1191.54ms | tok/sec: 880,016 | bf16_mfu: 54.06 | total time: 29.00m | eta: 82.1m step 01461/05568 (26.24%) | loss: 2.713036 | lrm: 1.00 | dt: 1196.17ms | tok/sec: 876,612 | bf16_mfu: 53.85 | total time: 29.02m | eta: 82.1m step 01462/05568 (26.26%) | loss: 2.709601 | lrm: 1.00 | dt: 1196.06ms | tok/sec: 876,688 | bf16_mfu: 53.85 | total time: 29.04m | eta: 82.1m step 01463/05568 (26.28%) | loss: 2.719277 | lrm: 1.00 | dt: 1195.85ms | tok/sec: 876,843 | bf16_mfu: 53.86 | total time: 29.06m | eta: 82.1m step 01464/05568 (26.29%) | loss: 2.718230 | lrm: 1.00 | dt: 1189.55ms | tok/sec: 881,489 | bf16_mfu: 54.15 | total time: 29.08m | eta: 82.1m step 01465/05568 (26.31%) | loss: 2.737818 | lrm: 1.00 | dt: 1197.33ms | tok/sec: 875,760 | bf16_mfu: 53.80 | total time: 29.10m | eta: 82.0m step 01466/05568 (26.33%) | loss: 2.728231 | lrm: 1.00 | dt: 1196.01ms | tok/sec: 876,727 | bf16_mfu: 53.86 | total time: 29.12m | eta: 82.0m step 01467/05568 (26.35%) | loss: 2.724464 | lrm: 1.00 | dt: 1192.34ms | tok/sec: 879,428 | bf16_mfu: 54.02 | total time: 29.14m | eta: 82.0m step 01468/05568 (26.36%) | loss: 2.718295 | lrm: 1.00 | dt: 1196.15ms | tok/sec: 876,627 | bf16_mfu: 53.85 | total time: 29.16m | eta: 82.0m step 01469/05568 (26.38%) | loss: 2.727685 | lrm: 1.00 | dt: 1197.53ms | tok/sec: 875,613 | bf16_mfu: 53.79 | total time: 29.18m | eta: 82.0m step 01470/05568 (26.40%) | loss: 2.722579 | lrm: 1.00 | dt: 1196.10ms | tok/sec: 876,664 | bf16_mfu: 53.85 | total time: 29.20m | eta: 81.9m step 01471/05568 (26.42%) | loss: 2.717978 | lrm: 1.00 | dt: 1193.08ms | tok/sec: 878,884 | bf16_mfu: 53.99 | total time: 29.22m | eta: 81.9m step 01472/05568 (26.44%) | loss: 2.714907 | lrm: 1.00 | dt: 1189.26ms | tok/sec: 881,703 | bf16_mfu: 54.16 | total time: 29.24m | eta: 81.9m step 01473/05568 (26.45%) | loss: 2.716920 | lrm: 1.00 | dt: 1194.23ms | tok/sec: 878,037 | bf16_mfu: 53.94 | total time: 29.26m | eta: 81.9m step 01474/05568 (26.47%) | loss: 2.719693 | lrm: 1.00 | dt: 1193.26ms | tok/sec: 878,751 | bf16_mfu: 53.98 | total time: 29.28m | eta: 81.9m step 01475/05568 (26.49%) | loss: 2.708632 | lrm: 1.00 | dt: 1193.64ms | tok/sec: 878,466 | bf16_mfu: 53.96 | total time: 29.30m | eta: 81.8m step 01476/05568 (26.51%) | loss: 2.709488 | lrm: 1.00 | dt: 1200.20ms | tok/sec: 873,669 | bf16_mfu: 53.67 | total time: 29.32m | eta: 81.8m step 01477/05568 (26.53%) | loss: 2.718482 | lrm: 1.00 | dt: 1196.16ms | tok/sec: 876,621 | bf16_mfu: 53.85 | total time: 29.34m | eta: 81.8m step 01478/05568 (26.54%) | loss: 2.707647 | lrm: 1.00 | dt: 1194.78ms | tok/sec: 877,629 | bf16_mfu: 53.91 | total time: 29.36m | eta: 81.8m step 01479/05568 (26.56%) | loss: 2.713636 | lrm: 1.00 | dt: 1191.51ms | tok/sec: 880,037 | bf16_mfu: 54.06 | total time: 29.38m | eta: 81.8m step 01480/05568 (26.58%) | loss: 2.710207 | lrm: 1.00 | dt: 1188.76ms | tok/sec: 882,078 | bf16_mfu: 54.18 | total time: 29.40m | eta: 81.7m step 01481/05568 (26.60%) | loss: 2.692026 | lrm: 1.00 | dt: 1193.28ms | tok/sec: 878,731 | bf16_mfu: 53.98 | total time: 29.42m | eta: 81.7m step 01482/05568 (26.62%) | loss: 2.701249 | lrm: 1.00 | dt: 1192.69ms | tok/sec: 879,165 | bf16_mfu: 54.01 | total time: 29.44m | eta: 81.7m step 01483/05568 (26.63%) | loss: 2.700758 | lrm: 1.00 | dt: 1197.43ms | tok/sec: 875,689 | bf16_mfu: 53.79 | total time: 29.46m | eta: 81.7m step 01484/05568 (26.65%) | loss: 2.711310 | lrm: 1.00 | dt: 1188.42ms | tok/sec: 882,327 | bf16_mfu: 54.20 | total time: 29.48m | eta: 81.7m step 01485/05568 (26.67%) | loss: 2.715142 | lrm: 1.00 | dt: 1198.44ms | tok/sec: 874,951 | bf16_mfu: 53.75 | total time: 29.50m | eta: 81.6m step 01486/05568 (26.69%) | loss: 2.715644 | lrm: 1.00 | dt: 1191.55ms | tok/sec: 880,011 | bf16_mfu: 54.06 | total time: 29.52m | eta: 81.6m step 01487/05568 (26.71%) | loss: 2.719182 | lrm: 1.00 | dt: 1192.42ms | tok/sec: 879,370 | bf16_mfu: 54.02 | total time: 29.54m | eta: 81.6m step 01488/05568 (26.72%) | loss: 2.719024 | lrm: 1.00 | dt: 1188.23ms | tok/sec: 882,471 | bf16_mfu: 54.21 | total time: 29.56m | eta: 81.6m step 01489/05568 (26.74%) | loss: 2.721396 | lrm: 1.00 | dt: 1188.89ms | tok/sec: 881,975 | bf16_mfu: 54.18 | total time: 29.58m | eta: 81.6m step 01490/05568 (26.76%) | loss: 2.707513 | lrm: 1.00 | dt: 1193.10ms | tok/sec: 878,866 | bf16_mfu: 53.99 | total time: 29.60m | eta: 81.5m step 01491/05568 (26.78%) | loss: 2.705559 | lrm: 1.00 | dt: 1189.79ms | tok/sec: 881,314 | bf16_mfu: 54.14 | total time: 29.62m | eta: 81.5m step 01492/05568 (26.80%) | loss: 2.696806 | lrm: 1.00 | dt: 1191.83ms | tok/sec: 879,803 | bf16_mfu: 54.04 | total time: 29.64m | eta: 81.5m step 01493/05568 (26.81%) | loss: 2.697602 | lrm: 1.00 | dt: 1197.43ms | tok/sec: 875,690 | bf16_mfu: 53.79 | total time: 29.66m | eta: 81.5m step 01494/05568 (26.83%) | loss: 2.701891 | lrm: 1.00 | dt: 1194.86ms | tok/sec: 877,572 | bf16_mfu: 53.91 | total time: 29.68m | eta: 81.5m step 01495/05568 (26.85%) | loss: 2.715851 | lrm: 1.00 | dt: 1190.11ms | tok/sec: 881,076 | bf16_mfu: 54.12 | total time: 29.70m | eta: 81.4m step 01496/05568 (26.87%) | loss: 2.730860 | lrm: 1.00 | dt: 1197.68ms | tok/sec: 875,502 | bf16_mfu: 53.78 | total time: 29.72m | eta: 81.4m step 01497/05568 (26.89%) | loss: 2.743870 | lrm: 1.00 | dt: 1190.78ms | tok/sec: 880,582 | bf16_mfu: 54.09 | total time: 29.74m | eta: 81.4m step 01498/05568 (26.90%) | loss: 2.741731 | lrm: 1.00 | dt: 1197.58ms | tok/sec: 875,581 | bf16_mfu: 53.78 | total time: 29.76m | eta: 81.4m step 01499/05568 (26.92%) | loss: 2.735171 | lrm: 1.00 | dt: 1199.48ms | tok/sec: 874,192 | bf16_mfu: 53.70 | total time: 29.78m | eta: 81.4m step:1500/5568 val_bpb:0.822766 val_time:3.90s step 01500/05568 (26.94%) | loss: 2.737736 | lrm: 1.00 | dt: 1192.54ms | tok/sec: 879,278 | bf16_mfu: 54.01 | total time: 29.80m | eta: 81.3m step 01501/05568 (26.96%) | loss: 2.742040 | lrm: 1.00 | dt: 1191.85ms | tok/sec: 879,788 | bf16_mfu: 54.04 | total time: 29.82m | eta: 81.3m step 01502/05568 (26.98%) | loss: 2.745625 | lrm: 1.00 | dt: 1194.44ms | tok/sec: 877,881 | bf16_mfu: 53.93 | total time: 29.84m | eta: 81.3m step 01503/05568 (26.99%) | loss: 2.747126 | lrm: 1.00 | dt: 1193.25ms | tok/sec: 878,755 | bf16_mfu: 53.98 | total time: 29.86m | eta: 81.3m step 01504/05568 (27.01%) | loss: 2.730618 | lrm: 1.00 | dt: 1192.37ms | tok/sec: 879,406 | bf16_mfu: 54.02 | total time: 29.88m | eta: 81.2m step 01505/05568 (27.03%) | loss: 2.730003 | lrm: 1.00 | dt: 1189.92ms | tok/sec: 881,217 | bf16_mfu: 54.13 | total time: 29.90m | eta: 81.2m step 01506/05568 (27.05%) | loss: 2.722196 | lrm: 1.00 | dt: 1193.08ms | tok/sec: 878,882 | bf16_mfu: 53.99 | total time: 29.92m | eta: 81.2m step 01507/05568 (27.07%) | loss: 2.700608 | lrm: 1.00 | dt: 1187.98ms | tok/sec: 882,655 | bf16_mfu: 54.22 | total time: 29.94m | eta: 81.2m step 01508/05568 (27.08%) | loss: 2.701129 | lrm: 1.00 | dt: 1193.60ms | tok/sec: 878,497 | bf16_mfu: 53.96 | total time: 29.96m | eta: 81.2m step 01509/05568 (27.10%) | loss: 2.711558 | lrm: 1.00 | dt: 1188.15ms | tok/sec: 882,527 | bf16_mfu: 54.21 | total time: 29.98m | eta: 81.1m step 01510/05568 (27.12%) | loss: 2.712971 | lrm: 1.00 | dt: 1193.44ms | tok/sec: 878,616 | bf16_mfu: 53.97 | total time: 30.00m | eta: 81.1m step 01511/05568 (27.14%) | loss: 2.719785 | lrm: 1.00 | dt: 1194.27ms | tok/sec: 878,002 | bf16_mfu: 53.93 | total time: 30.02m | eta: 81.1m step 01512/05568 (27.16%) | loss: 2.734846 | lrm: 1.00 | dt: 1194.26ms | tok/sec: 878,009 | bf16_mfu: 53.93 | total time: 30.04m | eta: 81.1m step 01513/05568 (27.17%) | loss: 2.737201 | lrm: 1.00 | dt: 1193.97ms | tok/sec: 878,227 | bf16_mfu: 53.95 | total time: 30.06m | eta: 81.1m step 01514/05568 (27.19%) | loss: 2.732513 | lrm: 1.00 | dt: 1195.29ms | tok/sec: 877,254 | bf16_mfu: 53.89 | total time: 30.08m | eta: 81.0m step 01515/05568 (27.21%) | loss: 2.737765 | lrm: 1.00 | dt: 1192.43ms | tok/sec: 879,357 | bf16_mfu: 54.02 | total time: 30.09m | eta: 81.0m step 01516/05568 (27.23%) | loss: 2.730333 | lrm: 1.00 | dt: 1191.56ms | tok/sec: 879,999 | bf16_mfu: 54.06 | total time: 30.11m | eta: 81.0m step 01517/05568 (27.24%) | loss: 2.738270 | lrm: 1.00 | dt: 1195.65ms | tok/sec: 876,992 | bf16_mfu: 53.87 | total time: 30.13m | eta: 81.0m step 01518/05568 (27.26%) | loss: 2.729146 | lrm: 1.00 | dt: 1190.21ms | tok/sec: 880,998 | bf16_mfu: 54.12 | total time: 30.15m | eta: 81.0m step 01519/05568 (27.28%) | loss: 2.726184 | lrm: 1.00 | dt: 1190.06ms | tok/sec: 881,111 | bf16_mfu: 54.12 | total time: 30.17m | eta: 80.9m step 01520/05568 (27.30%) | loss: 2.727291 | lrm: 1.00 | dt: 1195.62ms | tok/sec: 877,012 | bf16_mfu: 53.87 | total time: 30.19m | eta: 80.9m step 01521/05568 (27.32%) | loss: 2.719852 | lrm: 1.00 | dt: 1191.58ms | tok/sec: 879,985 | bf16_mfu: 54.06 | total time: 30.21m | eta: 80.9m step 01522/05568 (27.33%) | loss: 2.729208 | lrm: 1.00 | dt: 1194.73ms | tok/sec: 877,669 | bf16_mfu: 53.91 | total time: 30.23m | eta: 80.9m step 01523/05568 (27.35%) | loss: 2.733972 | lrm: 1.00 | dt: 1192.89ms | tok/sec: 879,024 | bf16_mfu: 54.00 | total time: 30.25m | eta: 80.9m step 01524/05568 (27.37%) | loss: 2.733029 | lrm: 1.00 | dt: 1195.70ms | tok/sec: 876,958 | bf16_mfu: 53.87 | total time: 30.27m | eta: 80.8m step 01525/05568 (27.39%) | loss: 2.723566 | lrm: 1.00 | dt: 1189.88ms | tok/sec: 881,246 | bf16_mfu: 54.13 | total time: 30.29m | eta: 80.8m step 01526/05568 (27.41%) | loss: 2.725289 | lrm: 1.00 | dt: 1200.23ms | tok/sec: 873,647 | bf16_mfu: 53.67 | total time: 30.31m | eta: 80.8m step 01527/05568 (27.42%) | loss: 2.717638 | lrm: 1.00 | dt: 1192.82ms | tok/sec: 879,073 | bf16_mfu: 54.00 | total time: 30.33m | eta: 80.8m step 01528/05568 (27.44%) | loss: 2.710549 | lrm: 1.00 | dt: 1189.70ms | tok/sec: 881,374 | bf16_mfu: 54.14 | total time: 30.35m | eta: 80.8m step 01529/05568 (27.46%) | loss: 2.717014 | lrm: 1.00 | dt: 1190.95ms | tok/sec: 880,454 | bf16_mfu: 54.08 | total time: 30.37m | eta: 80.7m step 01530/05568 (27.48%) | loss: 2.709280 | lrm: 1.00 | dt: 1197.75ms | tok/sec: 875,455 | bf16_mfu: 53.78 | total time: 30.39m | eta: 80.7m step 01531/05568 (27.50%) | loss: 2.696305 | lrm: 1.00 | dt: 1190.94ms | tok/sec: 880,460 | bf16_mfu: 54.08 | total time: 30.41m | eta: 80.7m step 01532/05568 (27.51%) | loss: 2.677973 | lrm: 1.00 | dt: 1195.77ms | tok/sec: 876,902 | bf16_mfu: 53.87 | total time: 30.43m | eta: 80.7m step 01533/05568 (27.53%) | loss: 2.679738 | lrm: 1.00 | dt: 1193.00ms | tok/sec: 878,940 | bf16_mfu: 53.99 | total time: 30.45m | eta: 80.7m step 01534/05568 (27.55%) | loss: 2.683690 | lrm: 1.00 | dt: 1195.07ms | tok/sec: 877,419 | bf16_mfu: 53.90 | total time: 30.47m | eta: 80.6m step 01535/05568 (27.57%) | loss: 2.685083 | lrm: 1.00 | dt: 1195.16ms | tok/sec: 877,350 | bf16_mfu: 53.89 | total time: 30.49m | eta: 80.6m step 01536/05568 (27.59%) | loss: 2.683000 | lrm: 1.00 | dt: 1195.48ms | tok/sec: 877,118 | bf16_mfu: 53.88 | total time: 30.51m | eta: 80.6m step 01537/05568 (27.60%) | loss: 2.686610 | lrm: 1.00 | dt: 1193.48ms | tok/sec: 878,583 | bf16_mfu: 53.97 | total time: 30.53m | eta: 80.6m step 01538/05568 (27.62%) | loss: 2.687995 | lrm: 1.00 | dt: 1191.03ms | tok/sec: 880,395 | bf16_mfu: 54.08 | total time: 30.55m | eta: 80.6m step 01539/05568 (27.64%) | loss: 2.693517 | lrm: 1.00 | dt: 1192.44ms | tok/sec: 879,356 | bf16_mfu: 54.02 | total time: 30.57m | eta: 80.5m step 01540/05568 (27.66%) | loss: 2.685720 | lrm: 1.00 | dt: 1194.58ms | tok/sec: 877,779 | bf16_mfu: 53.92 | total time: 30.59m | eta: 80.5m step 01541/05568 (27.68%) | loss: 2.692746 | lrm: 1.00 | dt: 1194.91ms | tok/sec: 877,533 | bf16_mfu: 53.90 | total time: 30.61m | eta: 80.5m step 01542/05568 (27.69%) | loss: 2.695746 | lrm: 1.00 | dt: 1197.82ms | tok/sec: 875,401 | bf16_mfu: 53.77 | total time: 30.63m | eta: 80.5m step 01543/05568 (27.71%) | loss: 2.692840 | lrm: 1.00 | dt: 1193.81ms | tok/sec: 878,344 | bf16_mfu: 53.95 | total time: 30.65m | eta: 80.5m step 01544/05568 (27.73%) | loss: 2.688528 | lrm: 1.00 | dt: 1199.96ms | tok/sec: 873,839 | bf16_mfu: 53.68 | total time: 30.67m | eta: 80.4m step 01545/05568 (27.75%) | loss: 2.682818 | lrm: 1.00 | dt: 1189.24ms | tok/sec: 881,722 | bf16_mfu: 54.16 | total time: 30.69m | eta: 80.4m step 01546/05568 (27.77%) | loss: 2.695279 | lrm: 1.00 | dt: 1194.23ms | tok/sec: 878,037 | bf16_mfu: 53.94 | total time: 30.71m | eta: 80.4m step 01547/05568 (27.78%) | loss: 2.694469 | lrm: 1.00 | dt: 1191.85ms | tok/sec: 879,788 | bf16_mfu: 54.04 | total time: 30.73m | eta: 80.4m step 01548/05568 (27.80%) | loss: 2.696735 | lrm: 1.00 | dt: 1191.74ms | tok/sec: 879,868 | bf16_mfu: 54.05 | total time: 30.75m | eta: 80.4m step 01549/05568 (27.82%) | loss: 2.696653 | lrm: 1.00 | dt: 1197.24ms | tok/sec: 875,826 | bf16_mfu: 53.80 | total time: 30.77m | eta: 80.3m step 01550/05568 (27.84%) | loss: 2.694838 | lrm: 1.00 | dt: 1194.60ms | tok/sec: 877,766 | bf16_mfu: 53.92 | total time: 30.79m | eta: 80.3m step 01551/05568 (27.86%) | loss: 2.692092 | lrm: 1.00 | dt: 1191.20ms | tok/sec: 880,271 | bf16_mfu: 54.07 | total time: 30.81m | eta: 80.3m step 01552/05568 (27.87%) | loss: 2.686663 | lrm: 1.00 | dt: 1197.09ms | tok/sec: 875,939 | bf16_mfu: 53.81 | total time: 30.83m | eta: 80.3m step 01553/05568 (27.89%) | loss: 2.680951 | lrm: 1.00 | dt: 1195.25ms | tok/sec: 877,285 | bf16_mfu: 53.89 | total time: 30.85m | eta: 80.3m step 01554/05568 (27.91%) | loss: 2.678074 | lrm: 1.00 | dt: 1194.34ms | tok/sec: 877,956 | bf16_mfu: 53.93 | total time: 30.87m | eta: 80.2m step 01555/05568 (27.93%) | loss: 2.676614 | lrm: 1.00 | dt: 1190.37ms | tok/sec: 880,881 | bf16_mfu: 54.11 | total time: 30.89m | eta: 80.2m step 01556/05568 (27.95%) | loss: 2.676597 | lrm: 1.00 | dt: 1193.98ms | tok/sec: 878,218 | bf16_mfu: 53.95 | total time: 30.91m | eta: 80.2m step 01557/05568 (27.96%) | loss: 2.668647 | lrm: 1.00 | dt: 1190.45ms | tok/sec: 880,820 | bf16_mfu: 54.11 | total time: 30.93m | eta: 80.2m step 01558/05568 (27.98%) | loss: 2.669982 | lrm: 1.00 | dt: 1193.98ms | tok/sec: 878,221 | bf16_mfu: 53.95 | total time: 30.95m | eta: 80.2m step 01559/05568 (28.00%) | loss: 2.667924 | lrm: 1.00 | dt: 1197.89ms | tok/sec: 875,354 | bf16_mfu: 53.77 | total time: 30.97m | eta: 80.1m step 01560/05568 (28.02%) | loss: 2.668095 | lrm: 1.00 | dt: 1193.52ms | tok/sec: 878,556 | bf16_mfu: 53.97 | total time: 30.99m | eta: 80.1m step 01561/05568 (28.04%) | loss: 2.666752 | lrm: 1.00 | dt: 1195.40ms | tok/sec: 877,176 | bf16_mfu: 53.88 | total time: 31.01m | eta: 80.1m step 01562/05568 (28.05%) | loss: 2.671270 | lrm: 1.00 | dt: 1195.99ms | tok/sec: 876,740 | bf16_mfu: 53.86 | total time: 31.03m | eta: 80.1m step 01563/05568 (28.07%) | loss: 2.667475 | lrm: 1.00 | dt: 1192.28ms | tok/sec: 879,470 | bf16_mfu: 54.02 | total time: 31.05m | eta: 80.1m step 01564/05568 (28.09%) | loss: 2.670894 | lrm: 1.00 | dt: 1194.89ms | tok/sec: 877,550 | bf16_mfu: 53.91 | total time: 31.07m | eta: 80.0m step 01565/05568 (28.11%) | loss: 2.677715 | lrm: 1.00 | dt: 1192.50ms | tok/sec: 879,310 | bf16_mfu: 54.01 | total time: 31.09m | eta: 80.0m step 01566/05568 (28.12%) | loss: 2.673228 | lrm: 1.00 | dt: 1196.68ms | tok/sec: 876,235 | bf16_mfu: 53.83 | total time: 31.11m | eta: 80.0m step 01567/05568 (28.14%) | loss: 2.678542 | lrm: 1.00 | dt: 1191.86ms | tok/sec: 879,779 | bf16_mfu: 54.04 | total time: 31.13m | eta: 80.0m step 01568/05568 (28.16%) | loss: 2.675013 | lrm: 1.00 | dt: 1188.20ms | tok/sec: 882,492 | bf16_mfu: 54.21 | total time: 31.15m | eta: 80.0m step 01569/05568 (28.18%) | loss: 2.673497 | lrm: 1.00 | dt: 1193.78ms | tok/sec: 878,366 | bf16_mfu: 53.96 | total time: 31.17m | eta: 79.9m step 01570/05568 (28.20%) | loss: 2.689138 | lrm: 1.00 | dt: 1187.27ms | tok/sec: 883,184 | bf16_mfu: 54.25 | total time: 31.19m | eta: 79.9m step 01571/05568 (28.21%) | loss: 2.675160 | lrm: 1.00 | dt: 1196.86ms | tok/sec: 876,108 | bf16_mfu: 53.82 | total time: 31.21m | eta: 79.9m step 01572/05568 (28.23%) | loss: 2.685602 | lrm: 1.00 | dt: 1191.68ms | tok/sec: 879,914 | bf16_mfu: 54.05 | total time: 31.23m | eta: 79.9m step 01573/05568 (28.25%) | loss: 2.682895 | lrm: 1.00 | dt: 1199.67ms | tok/sec: 874,052 | bf16_mfu: 53.69 | total time: 31.25m | eta: 79.9m step 01574/05568 (28.27%) | loss: 2.673631 | lrm: 1.00 | dt: 1194.18ms | tok/sec: 878,069 | bf16_mfu: 53.94 | total time: 31.27m | eta: 79.8m step 01575/05568 (28.29%) | loss: 2.677507 | lrm: 1.00 | dt: 1190.38ms | tok/sec: 880,871 | bf16_mfu: 54.11 | total time: 31.29m | eta: 79.8m step 01576/05568 (28.30%) | loss: 2.670510 | lrm: 1.00 | dt: 1191.40ms | tok/sec: 880,123 | bf16_mfu: 54.06 | total time: 31.31m | eta: 79.8m step 01577/05568 (28.32%) | loss: 2.668726 | lrm: 1.00 | dt: 1190.15ms | tok/sec: 881,043 | bf16_mfu: 54.12 | total time: 31.33m | eta: 79.8m step 01578/05568 (28.34%) | loss: 2.677698 | lrm: 1.00 | dt: 1195.62ms | tok/sec: 877,017 | bf16_mfu: 53.87 | total time: 31.35m | eta: 79.7m step 01579/05568 (28.36%) | loss: 2.691154 | lrm: 1.00 | dt: 1197.23ms | tok/sec: 875,836 | bf16_mfu: 53.80 | total time: 31.37m | eta: 79.7m step 01580/05568 (28.38%) | loss: 2.675191 | lrm: 1.00 | dt: 1186.79ms | tok/sec: 883,537 | bf16_mfu: 54.27 | total time: 31.39m | eta: 79.7m step 01581/05568 (28.39%) | loss: 2.685481 | lrm: 1.00 | dt: 1196.18ms | tok/sec: 876,603 | bf16_mfu: 53.85 | total time: 31.41m | eta: 79.7m step 01582/05568 (28.41%) | loss: 2.682431 | lrm: 1.00 | dt: 1196.09ms | tok/sec: 876,672 | bf16_mfu: 53.85 | total time: 31.43m | eta: 79.7m step 01583/05568 (28.43%) | loss: 2.679732 | lrm: 1.00 | dt: 1190.15ms | tok/sec: 881,044 | bf16_mfu: 54.12 | total time: 31.45m | eta: 79.6m step 01584/05568 (28.45%) | loss: 2.675646 | lrm: 1.00 | dt: 1193.88ms | tok/sec: 878,295 | bf16_mfu: 53.95 | total time: 31.47m | eta: 79.6m step 01585/05568 (28.47%) | loss: 2.676012 | lrm: 1.00 | dt: 1192.04ms | tok/sec: 879,651 | bf16_mfu: 54.03 | total time: 31.49m | eta: 79.6m step 01586/05568 (28.48%) | loss: 2.671107 | lrm: 1.00 | dt: 1194.61ms | tok/sec: 877,757 | bf16_mfu: 53.92 | total time: 31.51m | eta: 79.6m step 01587/05568 (28.50%) | loss: 2.675245 | lrm: 1.00 | dt: 1194.42ms | tok/sec: 877,897 | bf16_mfu: 53.93 | total time: 31.53m | eta: 79.6m step 01588/05568 (28.52%) | loss: 2.684679 | lrm: 1.00 | dt: 1193.47ms | tok/sec: 878,591 | bf16_mfu: 53.97 | total time: 31.55m | eta: 79.5m step 01589/05568 (28.54%) | loss: 2.691125 | lrm: 1.00 | dt: 1191.27ms | tok/sec: 880,217 | bf16_mfu: 54.07 | total time: 31.57m | eta: 79.5m step 01590/05568 (28.56%) | loss: 2.693628 | lrm: 1.00 | dt: 1195.30ms | tok/sec: 877,250 | bf16_mfu: 53.89 | total time: 31.59m | eta: 79.5m step 01591/05568 (28.57%) | loss: 2.686405 | lrm: 1.00 | dt: 1191.06ms | tok/sec: 880,371 | bf16_mfu: 54.08 | total time: 31.61m | eta: 79.5m step 01592/05568 (28.59%) | loss: 2.689122 | lrm: 1.00 | dt: 1192.63ms | tok/sec: 879,209 | bf16_mfu: 54.01 | total time: 31.63m | eta: 79.5m step 01593/05568 (28.61%) | loss: 2.680637 | lrm: 1.00 | dt: 1196.05ms | tok/sec: 876,696 | bf16_mfu: 53.85 | total time: 31.65m | eta: 79.4m step 01594/05568 (28.63%) | loss: 2.688756 | lrm: 1.00 | dt: 1192.92ms | tok/sec: 878,998 | bf16_mfu: 53.99 | total time: 31.67m | eta: 79.4m step 01595/05568 (28.65%) | loss: 2.686024 | lrm: 1.00 | dt: 1190.70ms | tok/sec: 880,637 | bf16_mfu: 54.10 | total time: 31.69m | eta: 79.4m step 01596/05568 (28.66%) | loss: 2.693564 | lrm: 1.00 | dt: 1194.57ms | tok/sec: 877,788 | bf16_mfu: 53.92 | total time: 31.71m | eta: 79.4m step 01597/05568 (28.68%) | loss: 2.692984 | lrm: 1.00 | dt: 1196.89ms | tok/sec: 876,080 | bf16_mfu: 53.82 | total time: 31.73m | eta: 79.4m step 01598/05568 (28.70%) | loss: 2.698234 | lrm: 1.00 | dt: 1191.52ms | tok/sec: 880,031 | bf16_mfu: 54.06 | total time: 31.75m | eta: 79.3m step 01599/05568 (28.72%) | loss: 2.698981 | lrm: 1.00 | dt: 1190.78ms | tok/sec: 880,576 | bf16_mfu: 54.09 | total time: 31.77m | eta: 79.3m step 01600/05568 (28.74%) | loss: 2.705399 | lrm: 1.00 | dt: 1194.90ms | tok/sec: 877,539 | bf16_mfu: 53.91 | total time: 31.79m | eta: 79.3m step 01601/05568 (28.75%) | loss: 2.700221 | lrm: 1.00 | dt: 1193.62ms | tok/sec: 878,483 | bf16_mfu: 53.96 | total time: 31.81m | eta: 79.3m step 01602/05568 (28.77%) | loss: 2.699838 | lrm: 1.00 | dt: 1195.30ms | tok/sec: 877,252 | bf16_mfu: 53.89 | total time: 31.83m | eta: 79.3m step 01603/05568 (28.79%) | loss: 2.690570 | lrm: 1.00 | dt: 1189.79ms | tok/sec: 881,311 | bf16_mfu: 54.14 | total time: 31.85m | eta: 79.2m step 01604/05568 (28.81%) | loss: 2.692063 | lrm: 1.00 | dt: 1189.85ms | tok/sec: 881,266 | bf16_mfu: 54.13 | total time: 31.87m | eta: 79.2m step 01605/05568 (28.83%) | loss: 2.685166 | lrm: 1.00 | dt: 1195.50ms | tok/sec: 877,099 | bf16_mfu: 53.88 | total time: 31.89m | eta: 79.2m step 01606/05568 (28.84%) | loss: 2.681345 | lrm: 1.00 | dt: 1197.85ms | tok/sec: 875,382 | bf16_mfu: 53.77 | total time: 31.91m | eta: 79.2m step 01607/05568 (28.86%) | loss: 2.692640 | lrm: 1.00 | dt: 1197.63ms | tok/sec: 875,539 | bf16_mfu: 53.78 | total time: 31.93m | eta: 79.2m step 01608/05568 (28.88%) | loss: 2.699646 | lrm: 1.00 | dt: 1191.46ms | tok/sec: 880,078 | bf16_mfu: 54.06 | total time: 31.94m | eta: 79.1m step 01609/05568 (28.90%) | loss: 2.698672 | lrm: 1.00 | dt: 1193.55ms | tok/sec: 878,538 | bf16_mfu: 53.97 | total time: 31.96m | eta: 79.1m step 01610/05568 (28.92%) | loss: 2.705287 | lrm: 1.00 | dt: 1193.99ms | tok/sec: 878,211 | bf16_mfu: 53.95 | total time: 31.98m | eta: 79.1m step 01611/05568 (28.93%) | loss: 2.695262 | lrm: 1.00 | dt: 1189.54ms | tok/sec: 881,497 | bf16_mfu: 54.15 | total time: 32.00m | eta: 79.1m step 01612/05568 (28.95%) | loss: 2.696283 | lrm: 1.00 | dt: 1193.63ms | tok/sec: 878,477 | bf16_mfu: 53.96 | total time: 32.02m | eta: 79.1m step 01613/05568 (28.97%) | loss: 2.696616 | lrm: 1.00 | dt: 1190.08ms | tok/sec: 881,093 | bf16_mfu: 54.12 | total time: 32.04m | eta: 79.0m step 01614/05568 (28.99%) | loss: 2.682776 | lrm: 1.00 | dt: 1194.11ms | tok/sec: 878,123 | bf16_mfu: 53.94 | total time: 32.06m | eta: 79.0m step 01615/05568 (29.01%) | loss: 2.687296 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,310 | bf16_mfu: 53.95 | total time: 32.08m | eta: 79.0m step 01616/05568 (29.02%) | loss: 2.682271 | lrm: 1.00 | dt: 1191.20ms | tok/sec: 880,271 | bf16_mfu: 54.07 | total time: 32.10m | eta: 79.0m step 01617/05568 (29.04%) | loss: 2.691099 | lrm: 1.00 | dt: 1191.53ms | tok/sec: 880,023 | bf16_mfu: 54.06 | total time: 32.12m | eta: 79.0m step 01618/05568 (29.06%) | loss: 2.700970 | lrm: 1.00 | dt: 1193.75ms | tok/sec: 878,390 | bf16_mfu: 53.96 | total time: 32.14m | eta: 78.9m step 01619/05568 (29.08%) | loss: 2.704899 | lrm: 1.00 | dt: 1192.61ms | tok/sec: 879,230 | bf16_mfu: 54.01 | total time: 32.16m | eta: 78.9m step 01620/05568 (29.09%) | loss: 2.688959 | lrm: 1.00 | dt: 1199.01ms | tok/sec: 874,533 | bf16_mfu: 53.72 | total time: 32.18m | eta: 78.9m step 01621/05568 (29.11%) | loss: 2.696158 | lrm: 1.00 | dt: 1199.92ms | tok/sec: 873,871 | bf16_mfu: 53.68 | total time: 32.20m | eta: 78.9m step 01622/05568 (29.13%) | loss: 2.707221 | lrm: 1.00 | dt: 1195.22ms | tok/sec: 877,310 | bf16_mfu: 53.89 | total time: 32.22m | eta: 78.9m step 01623/05568 (29.15%) | loss: 2.718927 | lrm: 1.00 | dt: 1195.21ms | tok/sec: 877,318 | bf16_mfu: 53.89 | total time: 32.24m | eta: 78.8m step 01624/05568 (29.17%) | loss: 2.734296 | lrm: 1.00 | dt: 1192.60ms | tok/sec: 879,238 | bf16_mfu: 54.01 | total time: 32.26m | eta: 78.8m step 01625/05568 (29.18%) | loss: 2.737584 | lrm: 1.00 | dt: 1192.94ms | tok/sec: 878,986 | bf16_mfu: 53.99 | total time: 32.28m | eta: 78.8m step 01626/05568 (29.20%) | loss: 2.732419 | lrm: 1.00 | dt: 1195.40ms | tok/sec: 877,176 | bf16_mfu: 53.88 | total time: 32.30m | eta: 78.8m step 01627/05568 (29.22%) | loss: 2.740070 | lrm: 1.00 | dt: 1193.09ms | tok/sec: 878,871 | bf16_mfu: 53.99 | total time: 32.32m | eta: 78.8m step 01628/05568 (29.24%) | loss: 2.737420 | lrm: 1.00 | dt: 1191.00ms | tok/sec: 880,418 | bf16_mfu: 54.08 | total time: 32.34m | eta: 78.7m step 01629/05568 (29.26%) | loss: 2.738059 | lrm: 1.00 | dt: 1194.06ms | tok/sec: 878,158 | bf16_mfu: 53.94 | total time: 32.36m | eta: 78.7m step 01630/05568 (29.27%) | loss: 2.747711 | lrm: 1.00 | dt: 1190.07ms | tok/sec: 881,100 | bf16_mfu: 54.12 | total time: 32.38m | eta: 78.7m step 01631/05568 (29.29%) | loss: 2.746065 | lrm: 1.00 | dt: 1191.00ms | tok/sec: 880,413 | bf16_mfu: 54.08 | total time: 32.40m | eta: 78.7m step 01632/05568 (29.31%) | loss: 2.742689 | lrm: 1.00 | dt: 1196.85ms | tok/sec: 876,114 | bf16_mfu: 53.82 | total time: 32.42m | eta: 78.7m step 01633/05568 (29.33%) | loss: 2.731016 | lrm: 1.00 | dt: 1195.14ms | tok/sec: 877,366 | bf16_mfu: 53.89 | total time: 32.44m | eta: 78.6m step 01634/05568 (29.35%) | loss: 2.732289 | lrm: 1.00 | dt: 1197.28ms | tok/sec: 875,797 | bf16_mfu: 53.80 | total time: 32.46m | eta: 78.6m step 01635/05568 (29.36%) | loss: 2.722794 | lrm: 1.00 | dt: 1192.50ms | tok/sec: 879,312 | bf16_mfu: 54.01 | total time: 32.48m | eta: 78.6m step 01636/05568 (29.38%) | loss: 2.722186 | lrm: 1.00 | dt: 1198.89ms | tok/sec: 874,619 | bf16_mfu: 53.73 | total time: 32.50m | eta: 78.6m step 01637/05568 (29.40%) | loss: 2.718619 | lrm: 1.00 | dt: 1193.51ms | tok/sec: 878,562 | bf16_mfu: 53.97 | total time: 32.52m | eta: 78.6m step 01638/05568 (29.42%) | loss: 2.711006 | lrm: 1.00 | dt: 1192.31ms | tok/sec: 879,452 | bf16_mfu: 54.02 | total time: 32.54m | eta: 78.5m step 01639/05568 (29.44%) | loss: 2.699775 | lrm: 1.00 | dt: 1195.28ms | tok/sec: 877,265 | bf16_mfu: 53.89 | total time: 32.56m | eta: 78.5m step 01640/05568 (29.45%) | loss: 2.704471 | lrm: 1.00 | dt: 1191.53ms | tok/sec: 880,027 | bf16_mfu: 54.06 | total time: 32.58m | eta: 78.5m step 01641/05568 (29.47%) | loss: 2.716881 | lrm: 1.00 | dt: 1195.25ms | tok/sec: 877,287 | bf16_mfu: 53.89 | total time: 32.60m | eta: 78.5m step 01642/05568 (29.49%) | loss: 2.705602 | lrm: 1.00 | dt: 1194.11ms | tok/sec: 878,121 | bf16_mfu: 53.94 | total time: 32.62m | eta: 78.5m step 01643/05568 (29.51%) | loss: 2.694012 | lrm: 1.00 | dt: 1195.24ms | tok/sec: 877,294 | bf16_mfu: 53.89 | total time: 32.64m | eta: 78.4m step 01644/05568 (29.53%) | loss: 2.702662 | lrm: 1.00 | dt: 1195.89ms | tok/sec: 876,813 | bf16_mfu: 53.86 | total time: 32.66m | eta: 78.4m step 01645/05568 (29.54%) | loss: 2.697392 | lrm: 1.00 | dt: 1194.08ms | tok/sec: 878,146 | bf16_mfu: 53.94 | total time: 32.68m | eta: 78.4m step 01646/05568 (29.56%) | loss: 2.705546 | lrm: 1.00 | dt: 1189.60ms | tok/sec: 881,451 | bf16_mfu: 54.15 | total time: 32.70m | eta: 78.4m step 01647/05568 (29.58%) | loss: 2.703513 | lrm: 1.00 | dt: 1198.07ms | tok/sec: 875,224 | bf16_mfu: 53.76 | total time: 32.72m | eta: 78.4m step 01648/05568 (29.60%) | loss: 2.690082 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,266 | bf16_mfu: 53.95 | total time: 32.74m | eta: 78.3m step 01649/05568 (29.62%) | loss: 2.690796 | lrm: 1.00 | dt: 1187.66ms | tok/sec: 882,895 | bf16_mfu: 54.23 | total time: 32.76m | eta: 78.3m step 01650/05568 (29.63%) | loss: 2.683989 | lrm: 1.00 | dt: 1192.91ms | tok/sec: 879,008 | bf16_mfu: 54.00 | total time: 32.78m | eta: 78.3m step 01651/05568 (29.65%) | loss: 2.685786 | lrm: 1.00 | dt: 1188.29ms | tok/sec: 882,420 | bf16_mfu: 54.21 | total time: 32.80m | eta: 78.3m step 01652/05568 (29.67%) | loss: 2.694442 | lrm: 1.00 | dt: 1194.26ms | tok/sec: 878,010 | bf16_mfu: 53.93 | total time: 32.82m | eta: 78.3m step 01653/05568 (29.69%) | loss: 2.684027 | lrm: 1.00 | dt: 1191.70ms | tok/sec: 879,896 | bf16_mfu: 54.05 | total time: 32.84m | eta: 78.2m step 01654/05568 (29.71%) | loss: 2.684190 | lrm: 1.00 | dt: 1191.02ms | tok/sec: 880,404 | bf16_mfu: 54.08 | total time: 32.86m | eta: 78.2m step 01655/05568 (29.72%) | loss: 2.691166 | lrm: 1.00 | dt: 1193.05ms | tok/sec: 878,904 | bf16_mfu: 53.99 | total time: 32.88m | eta: 78.2m step 01656/05568 (29.74%) | loss: 2.692542 | lrm: 1.00 | dt: 1200.63ms | tok/sec: 873,352 | bf16_mfu: 53.65 | total time: 32.90m | eta: 78.2m step 01657/05568 (29.76%) | loss: 2.681784 | lrm: 1.00 | dt: 1190.85ms | tok/sec: 880,527 | bf16_mfu: 54.09 | total time: 32.92m | eta: 78.2m step 01658/05568 (29.78%) | loss: 2.681458 | lrm: 1.00 | dt: 1193.95ms | tok/sec: 878,242 | bf16_mfu: 53.95 | total time: 32.94m | eta: 78.1m step 01659/05568 (29.80%) | loss: 2.674404 | lrm: 1.00 | dt: 1192.03ms | tok/sec: 879,655 | bf16_mfu: 54.04 | total time: 32.96m | eta: 78.1m step 01660/05568 (29.81%) | loss: 2.672671 | lrm: 1.00 | dt: 1187.16ms | tok/sec: 883,262 | bf16_mfu: 54.26 | total time: 32.98m | eta: 78.1m step 01661/05568 (29.83%) | loss: 2.679837 | lrm: 1.00 | dt: 1195.45ms | tok/sec: 877,137 | bf16_mfu: 53.88 | total time: 33.00m | eta: 78.1m step 01662/05568 (29.85%) | loss: 2.665815 | lrm: 1.00 | dt: 1191.26ms | tok/sec: 880,226 | bf16_mfu: 54.07 | total time: 33.02m | eta: 78.1m step 01663/05568 (29.87%) | loss: 2.669564 | lrm: 1.00 | dt: 1193.05ms | tok/sec: 878,903 | bf16_mfu: 53.99 | total time: 33.04m | eta: 78.0m step 01664/05568 (29.89%) | loss: 2.674739 | lrm: 1.00 | dt: 1195.48ms | tok/sec: 877,119 | bf16_mfu: 53.88 | total time: 33.06m | eta: 78.0m step 01665/05568 (29.90%) | loss: 2.665638 | lrm: 1.00 | dt: 1193.33ms | tok/sec: 878,697 | bf16_mfu: 53.98 | total time: 33.08m | eta: 78.0m step 01666/05568 (29.92%) | loss: 2.692183 | lrm: 1.00 | dt: 1193.04ms | tok/sec: 878,911 | bf16_mfu: 53.99 | total time: 33.10m | eta: 78.0m step 01667/05568 (29.94%) | loss: 2.687608 | lrm: 1.00 | dt: 1189.78ms | tok/sec: 881,319 | bf16_mfu: 54.14 | total time: 33.12m | eta: 77.9m step 01668/05568 (29.96%) | loss: 2.697175 | lrm: 1.00 | dt: 1193.16ms | tok/sec: 878,819 | bf16_mfu: 53.98 | total time: 33.14m | eta: 77.9m step 01669/05568 (29.97%) | loss: 2.684653 | lrm: 1.00 | dt: 1195.63ms | tok/sec: 877,008 | bf16_mfu: 53.87 | total time: 33.16m | eta: 77.9m step 01670/05568 (29.99%) | loss: 2.686630 | lrm: 1.00 | dt: 1190.88ms | tok/sec: 880,505 | bf16_mfu: 54.09 | total time: 33.18m | eta: 77.9m step 01671/05568 (30.01%) | loss: 2.693395 | lrm: 1.00 | dt: 1191.54ms | tok/sec: 880,013 | bf16_mfu: 54.06 | total time: 33.20m | eta: 77.9m step 01672/05568 (30.03%) | loss: 2.700655 | lrm: 1.00 | dt: 1192.54ms | tok/sec: 879,279 | bf16_mfu: 54.01 | total time: 33.22m | eta: 77.8m step 01673/05568 (30.05%) | loss: 2.697002 | lrm: 1.00 | dt: 1192.43ms | tok/sec: 879,358 | bf16_mfu: 54.02 | total time: 33.24m | eta: 77.8m step 01674/05568 (30.06%) | loss: 2.697236 | lrm: 1.00 | dt: 1192.40ms | tok/sec: 879,384 | bf16_mfu: 54.02 | total time: 33.26m | eta: 77.8m step 01675/05568 (30.08%) | loss: 2.703806 | lrm: 1.00 | dt: 1190.84ms | tok/sec: 880,533 | bf16_mfu: 54.09 | total time: 33.28m | eta: 77.8m step 01676/05568 (30.10%) | loss: 2.695182 | lrm: 1.00 | dt: 1192.70ms | tok/sec: 879,162 | bf16_mfu: 54.00 | total time: 33.30m | eta: 77.8m step 01677/05568 (30.12%) | loss: 2.689659 | lrm: 1.00 | dt: 1194.90ms | tok/sec: 877,539 | bf16_mfu: 53.91 | total time: 33.32m | eta: 77.7m step 01678/05568 (30.14%) | loss: 2.682146 | lrm: 1.00 | dt: 1194.34ms | tok/sec: 877,953 | bf16_mfu: 53.93 | total time: 33.34m | eta: 77.7m step 01679/05568 (30.15%) | loss: 2.692335 | lrm: 1.00 | dt: 1192.05ms | tok/sec: 879,638 | bf16_mfu: 54.03 | total time: 33.36m | eta: 77.7m step 01680/05568 (30.17%) | loss: 2.691309 | lrm: 1.00 | dt: 1191.05ms | tok/sec: 880,377 | bf16_mfu: 54.08 | total time: 33.38m | eta: 77.7m step 01681/05568 (30.19%) | loss: 2.697356 | lrm: 1.00 | dt: 1193.55ms | tok/sec: 878,533 | bf16_mfu: 53.97 | total time: 33.40m | eta: 77.7m step 01682/05568 (30.21%) | loss: 2.699372 | lrm: 1.00 | dt: 1198.62ms | tok/sec: 874,822 | bf16_mfu: 53.74 | total time: 33.42m | eta: 77.6m step 01683/05568 (30.23%) | loss: 2.703595 | lrm: 1.00 | dt: 1197.53ms | tok/sec: 875,616 | bf16_mfu: 53.79 | total time: 33.44m | eta: 77.6m step 01684/05568 (30.24%) | loss: 2.702352 | lrm: 1.00 | dt: 1192.55ms | tok/sec: 879,270 | bf16_mfu: 54.01 | total time: 33.46m | eta: 77.6m step 01685/05568 (30.26%) | loss: 2.702612 | lrm: 1.00 | dt: 1192.54ms | tok/sec: 879,282 | bf16_mfu: 54.01 | total time: 33.48m | eta: 77.6m step 01686/05568 (30.28%) | loss: 2.696901 | lrm: 1.00 | dt: 1193.14ms | tok/sec: 878,837 | bf16_mfu: 53.98 | total time: 33.50m | eta: 77.6m step 01687/05568 (30.30%) | loss: 2.683825 | lrm: 1.00 | dt: 1193.21ms | tok/sec: 878,788 | bf16_mfu: 53.98 | total time: 33.52m | eta: 77.5m step 01688/05568 (30.32%) | loss: 2.683391 | lrm: 1.00 | dt: 1191.06ms | tok/sec: 880,370 | bf16_mfu: 54.08 | total time: 33.54m | eta: 77.5m step 01689/05568 (30.33%) | loss: 2.688391 | lrm: 1.00 | dt: 1188.56ms | tok/sec: 882,222 | bf16_mfu: 54.19 | total time: 33.56m | eta: 77.5m step 01690/05568 (30.35%) | loss: 2.687621 | lrm: 1.00 | dt: 1194.79ms | tok/sec: 877,625 | bf16_mfu: 53.91 | total time: 33.58m | eta: 77.5m step 01691/05568 (30.37%) | loss: 2.693205 | lrm: 1.00 | dt: 1189.94ms | tok/sec: 881,204 | bf16_mfu: 54.13 | total time: 33.60m | eta: 77.5m step 01692/05568 (30.39%) | loss: 2.701205 | lrm: 1.00 | dt: 1196.56ms | tok/sec: 876,324 | bf16_mfu: 53.83 | total time: 33.62m | eta: 77.4m step 01693/05568 (30.41%) | loss: 2.704749 | lrm: 1.00 | dt: 1191.97ms | tok/sec: 879,701 | bf16_mfu: 54.04 | total time: 33.64m | eta: 77.4m step 01694/05568 (30.42%) | loss: 2.718000 | lrm: 1.00 | dt: 1191.28ms | tok/sec: 880,209 | bf16_mfu: 54.07 | total time: 33.66m | eta: 77.4m step 01695/05568 (30.44%) | loss: 2.714656 | lrm: 1.00 | dt: 1193.84ms | tok/sec: 878,321 | bf16_mfu: 53.95 | total time: 33.68m | eta: 77.4m step 01696/05568 (30.46%) | loss: 2.717500 | lrm: 1.00 | dt: 1195.95ms | tok/sec: 876,770 | bf16_mfu: 53.86 | total time: 33.70m | eta: 77.4m step 01697/05568 (30.48%) | loss: 2.708331 | lrm: 1.00 | dt: 1188.33ms | tok/sec: 882,397 | bf16_mfu: 54.20 | total time: 33.71m | eta: 77.3m step 01698/05568 (30.50%) | loss: 2.707872 | lrm: 1.00 | dt: 1195.36ms | tok/sec: 877,202 | bf16_mfu: 53.88 | total time: 33.73m | eta: 77.3m step 01699/05568 (30.51%) | loss: 2.695843 | lrm: 1.00 | dt: 1188.28ms | tok/sec: 882,430 | bf16_mfu: 54.21 | total time: 33.75m | eta: 77.3m step 01700/05568 (30.53%) | loss: 2.691670 | lrm: 1.00 | dt: 1192.34ms | tok/sec: 879,425 | bf16_mfu: 54.02 | total time: 33.77m | eta: 77.3m step 01701/05568 (30.55%) | loss: 2.699761 | lrm: 1.00 | dt: 1194.76ms | tok/sec: 877,644 | bf16_mfu: 53.91 | total time: 33.79m | eta: 77.3m step 01702/05568 (30.57%) | loss: 2.704725 | lrm: 1.00 | dt: 1187.33ms | tok/sec: 883,135 | bf16_mfu: 54.25 | total time: 33.81m | eta: 77.2m step 01703/05568 (30.59%) | loss: 2.695472 | lrm: 1.00 | dt: 1193.95ms | tok/sec: 878,242 | bf16_mfu: 53.95 | total time: 33.83m | eta: 77.2m step 01704/05568 (30.60%) | loss: 2.688029 | lrm: 1.00 | dt: 1189.66ms | tok/sec: 881,405 | bf16_mfu: 54.14 | total time: 33.85m | eta: 77.2m step 01705/05568 (30.62%) | loss: 2.691072 | lrm: 1.00 | dt: 1196.21ms | tok/sec: 876,582 | bf16_mfu: 53.85 | total time: 33.87m | eta: 77.2m step 01706/05568 (30.64%) | loss: 2.697205 | lrm: 1.00 | dt: 1196.63ms | tok/sec: 876,276 | bf16_mfu: 53.83 | total time: 33.89m | eta: 77.2m step 01707/05568 (30.66%) | loss: 2.693078 | lrm: 1.00 | dt: 1196.57ms | tok/sec: 876,319 | bf16_mfu: 53.83 | total time: 33.91m | eta: 77.1m step 01708/05568 (30.68%) | loss: 2.696393 | lrm: 1.00 | dt: 1188.70ms | tok/sec: 882,119 | bf16_mfu: 54.19 | total time: 33.93m | eta: 77.1m step 01709/05568 (30.69%) | loss: 2.696001 | lrm: 1.00 | dt: 1192.75ms | tok/sec: 879,122 | bf16_mfu: 54.00 | total time: 33.95m | eta: 77.1m step 01710/05568 (30.71%) | loss: 2.700374 | lrm: 1.00 | dt: 1190.23ms | tok/sec: 880,985 | bf16_mfu: 54.12 | total time: 33.97m | eta: 77.1m step 01711/05568 (30.73%) | loss: 2.694663 | lrm: 1.00 | dt: 1192.38ms | tok/sec: 879,396 | bf16_mfu: 54.02 | total time: 33.99m | eta: 77.1m step 01712/05568 (30.75%) | loss: 2.692226 | lrm: 1.00 | dt: 1200.29ms | tok/sec: 873,600 | bf16_mfu: 53.66 | total time: 34.01m | eta: 77.0m step 01713/05568 (30.77%) | loss: 2.687884 | lrm: 1.00 | dt: 1194.95ms | tok/sec: 877,503 | bf16_mfu: 53.90 | total time: 34.03m | eta: 77.0m step 01714/05568 (30.78%) | loss: 2.694484 | lrm: 1.00 | dt: 1186.04ms | tok/sec: 884,101 | bf16_mfu: 54.31 | total time: 34.05m | eta: 77.0m step 01715/05568 (30.80%) | loss: 2.700381 | lrm: 1.00 | dt: 1193.03ms | tok/sec: 878,917 | bf16_mfu: 53.99 | total time: 34.07m | eta: 77.0m step 01716/05568 (30.82%) | loss: 2.704291 | lrm: 1.00 | dt: 1187.55ms | tok/sec: 882,972 | bf16_mfu: 54.24 | total time: 34.09m | eta: 77.0m step 01717/05568 (30.84%) | loss: 2.698836 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,635 | bf16_mfu: 53.91 | total time: 34.11m | eta: 76.9m step 01718/05568 (30.85%) | loss: 2.696851 | lrm: 1.00 | dt: 1197.09ms | tok/sec: 875,935 | bf16_mfu: 53.81 | total time: 34.13m | eta: 76.9m step 01719/05568 (30.87%) | loss: 2.696069 | lrm: 1.00 | dt: 1188.88ms | tok/sec: 881,986 | bf16_mfu: 54.18 | total time: 34.15m | eta: 76.9m step 01720/05568 (30.89%) | loss: 2.691131 | lrm: 1.00 | dt: 1197.26ms | tok/sec: 875,812 | bf16_mfu: 53.80 | total time: 34.17m | eta: 76.9m step 01721/05568 (30.91%) | loss: 2.700855 | lrm: 1.00 | dt: 1191.38ms | tok/sec: 880,133 | bf16_mfu: 54.06 | total time: 34.19m | eta: 76.9m step 01722/05568 (30.93%) | loss: 2.701526 | lrm: 1.00 | dt: 1195.53ms | tok/sec: 877,081 | bf16_mfu: 53.88 | total time: 34.21m | eta: 76.8m step 01723/05568 (30.94%) | loss: 2.698816 | lrm: 1.00 | dt: 1192.78ms | tok/sec: 879,099 | bf16_mfu: 54.00 | total time: 34.23m | eta: 76.8m step 01724/05568 (30.96%) | loss: 2.701867 | lrm: 1.00 | dt: 1189.82ms | tok/sec: 881,286 | bf16_mfu: 54.14 | total time: 34.25m | eta: 76.8m step 01725/05568 (30.98%) | loss: 2.704558 | lrm: 1.00 | dt: 1197.15ms | tok/sec: 875,895 | bf16_mfu: 53.80 | total time: 34.27m | eta: 76.8m step 01726/05568 (31.00%) | loss: 2.704976 | lrm: 1.00 | dt: 1191.31ms | tok/sec: 880,189 | bf16_mfu: 54.07 | total time: 34.29m | eta: 76.8m step 01727/05568 (31.02%) | loss: 2.707287 | lrm: 1.00 | dt: 1195.14ms | tok/sec: 877,363 | bf16_mfu: 53.89 | total time: 34.31m | eta: 76.7m step 01728/05568 (31.03%) | loss: 2.705133 | lrm: 1.00 | dt: 1194.56ms | tok/sec: 877,790 | bf16_mfu: 53.92 | total time: 34.33m | eta: 76.7m step 01729/05568 (31.05%) | loss: 2.712810 | lrm: 1.00 | dt: 1190.55ms | tok/sec: 880,752 | bf16_mfu: 54.10 | total time: 34.35m | eta: 76.7m step 01730/05568 (31.07%) | loss: 2.711461 | lrm: 1.00 | dt: 1200.36ms | tok/sec: 873,552 | bf16_mfu: 53.66 | total time: 34.37m | eta: 76.7m step 01731/05568 (31.09%) | loss: 2.706090 | lrm: 1.00 | dt: 1192.26ms | tok/sec: 879,486 | bf16_mfu: 54.02 | total time: 34.39m | eta: 76.7m step 01732/05568 (31.11%) | loss: 2.697947 | lrm: 1.00 | dt: 1188.84ms | tok/sec: 882,012 | bf16_mfu: 54.18 | total time: 34.41m | eta: 76.6m step 01733/05568 (31.12%) | loss: 2.703287 | lrm: 1.00 | dt: 1193.46ms | tok/sec: 878,598 | bf16_mfu: 53.97 | total time: 34.43m | eta: 76.6m step 01734/05568 (31.14%) | loss: 2.685865 | lrm: 1.00 | dt: 1196.68ms | tok/sec: 876,236 | bf16_mfu: 53.83 | total time: 34.45m | eta: 76.6m step 01735/05568 (31.16%) | loss: 2.701904 | lrm: 1.00 | dt: 1197.57ms | tok/sec: 875,582 | bf16_mfu: 53.79 | total time: 34.47m | eta: 76.6m step 01736/05568 (31.18%) | loss: 2.706165 | lrm: 1.00 | dt: 1188.99ms | tok/sec: 881,905 | bf16_mfu: 54.17 | total time: 34.49m | eta: 76.6m step 01737/05568 (31.20%) | loss: 2.701569 | lrm: 1.00 | dt: 1194.21ms | tok/sec: 878,049 | bf16_mfu: 53.94 | total time: 34.51m | eta: 76.5m step 01738/05568 (31.21%) | loss: 2.701305 | lrm: 1.00 | dt: 1195.94ms | tok/sec: 876,777 | bf16_mfu: 53.86 | total time: 34.53m | eta: 76.5m step 01739/05568 (31.23%) | loss: 2.707076 | lrm: 1.00 | dt: 1191.30ms | tok/sec: 880,194 | bf16_mfu: 54.07 | total time: 34.55m | eta: 76.5m step 01740/05568 (31.25%) | loss: 2.719263 | lrm: 1.00 | dt: 1196.22ms | tok/sec: 876,571 | bf16_mfu: 53.85 | total time: 34.57m | eta: 76.5m step 01741/05568 (31.27%) | loss: 2.722964 | lrm: 1.00 | dt: 1196.97ms | tok/sec: 876,028 | bf16_mfu: 53.81 | total time: 34.59m | eta: 76.5m step 01742/05568 (31.29%) | loss: 2.703416 | lrm: 1.00 | dt: 1192.68ms | tok/sec: 879,174 | bf16_mfu: 54.01 | total time: 34.61m | eta: 76.4m step 01743/05568 (31.30%) | loss: 2.699699 | lrm: 1.00 | dt: 1193.02ms | tok/sec: 878,924 | bf16_mfu: 53.99 | total time: 34.63m | eta: 76.4m step 01744/05568 (31.32%) | loss: 2.694287 | lrm: 1.00 | dt: 1195.62ms | tok/sec: 877,015 | bf16_mfu: 53.87 | total time: 34.65m | eta: 76.4m step 01745/05568 (31.34%) | loss: 2.694030 | lrm: 1.00 | dt: 1193.60ms | tok/sec: 878,499 | bf16_mfu: 53.96 | total time: 34.67m | eta: 76.4m step 01746/05568 (31.36%) | loss: 2.693084 | lrm: 1.00 | dt: 1193.79ms | tok/sec: 878,358 | bf16_mfu: 53.96 | total time: 34.69m | eta: 76.4m step 01747/05568 (31.38%) | loss: 2.696633 | lrm: 1.00 | dt: 1195.75ms | tok/sec: 876,916 | bf16_mfu: 53.87 | total time: 34.71m | eta: 76.3m step 01748/05568 (31.39%) | loss: 2.680441 | lrm: 1.00 | dt: 1193.73ms | tok/sec: 878,403 | bf16_mfu: 53.96 | total time: 34.73m | eta: 76.3m step 01749/05568 (31.41%) | loss: 2.670531 | lrm: 1.00 | dt: 1195.66ms | tok/sec: 876,981 | bf16_mfu: 53.87 | total time: 34.75m | eta: 76.3m step:1750/5568 val_bpb:0.815902 val_time:3.89s step 01750/05568 (31.43%) | loss: 2.674773 | lrm: 1.00 | dt: 1194.97ms | tok/sec: 877,493 | bf16_mfu: 53.90 | total time: 34.77m | eta: 76.3m step 01751/05568 (31.45%) | loss: 2.674285 | lrm: 1.00 | dt: 1194.42ms | tok/sec: 877,896 | bf16_mfu: 53.93 | total time: 34.79m | eta: 76.3m step 01752/05568 (31.47%) | loss: 2.676075 | lrm: 1.00 | dt: 1189.38ms | tok/sec: 881,618 | bf16_mfu: 54.16 | total time: 34.81m | eta: 76.2m step 01753/05568 (31.48%) | loss: 2.672548 | lrm: 1.00 | dt: 1194.27ms | tok/sec: 878,007 | bf16_mfu: 53.93 | total time: 34.83m | eta: 76.2m step 01754/05568 (31.50%) | loss: 2.683411 | lrm: 1.00 | dt: 1191.90ms | tok/sec: 879,754 | bf16_mfu: 54.04 | total time: 34.85m | eta: 76.2m step 01755/05568 (31.52%) | loss: 2.682792 | lrm: 1.00 | dt: 1195.71ms | tok/sec: 876,944 | bf16_mfu: 53.87 | total time: 34.87m | eta: 76.2m step 01756/05568 (31.54%) | loss: 2.664872 | lrm: 1.00 | dt: 1193.13ms | tok/sec: 878,843 | bf16_mfu: 53.99 | total time: 34.89m | eta: 76.2m step 01757/05568 (31.56%) | loss: 2.662605 | lrm: 1.00 | dt: 1191.70ms | tok/sec: 879,901 | bf16_mfu: 54.05 | total time: 34.91m | eta: 76.1m step 01758/05568 (31.57%) | loss: 2.664722 | lrm: 1.00 | dt: 1192.46ms | tok/sec: 879,339 | bf16_mfu: 54.02 | total time: 34.93m | eta: 76.1m step 01759/05568 (31.59%) | loss: 2.666759 | lrm: 1.00 | dt: 1194.62ms | tok/sec: 877,745 | bf16_mfu: 53.92 | total time: 34.95m | eta: 76.1m step 01760/05568 (31.61%) | loss: 2.665939 | lrm: 1.00 | dt: 1197.12ms | tok/sec: 875,914 | bf16_mfu: 53.81 | total time: 34.97m | eta: 76.1m step 01761/05568 (31.63%) | loss: 2.671711 | lrm: 1.00 | dt: 1191.79ms | tok/sec: 879,830 | bf16_mfu: 54.05 | total time: 34.99m | eta: 76.1m step 01762/05568 (31.65%) | loss: 2.668652 | lrm: 1.00 | dt: 1194.30ms | tok/sec: 877,984 | bf16_mfu: 53.93 | total time: 35.01m | eta: 76.0m step 01763/05568 (31.66%) | loss: 2.666279 | lrm: 1.00 | dt: 1197.97ms | tok/sec: 875,296 | bf16_mfu: 53.77 | total time: 35.03m | eta: 76.0m step 01764/05568 (31.68%) | loss: 2.662045 | lrm: 1.00 | dt: 1198.64ms | tok/sec: 874,808 | bf16_mfu: 53.74 | total time: 35.05m | eta: 76.0m step 01765/05568 (31.70%) | loss: 2.667773 | lrm: 1.00 | dt: 1199.26ms | tok/sec: 874,351 | bf16_mfu: 53.71 | total time: 35.07m | eta: 76.0m step 01766/05568 (31.72%) | loss: 2.668236 | lrm: 1.00 | dt: 1195.50ms | tok/sec: 877,101 | bf16_mfu: 53.88 | total time: 35.09m | eta: 76.0m step 01767/05568 (31.73%) | loss: 2.669032 | lrm: 1.00 | dt: 1192.09ms | tok/sec: 879,614 | bf16_mfu: 54.03 | total time: 35.11m | eta: 75.9m step 01768/05568 (31.75%) | loss: 2.672405 | lrm: 1.00 | dt: 1188.94ms | tok/sec: 881,938 | bf16_mfu: 54.18 | total time: 35.13m | eta: 75.9m step 01769/05568 (31.77%) | loss: 2.670179 | lrm: 1.00 | dt: 1196.81ms | tok/sec: 876,140 | bf16_mfu: 53.82 | total time: 35.15m | eta: 75.9m step 01770/05568 (31.79%) | loss: 2.674022 | lrm: 1.00 | dt: 1196.10ms | tok/sec: 876,665 | bf16_mfu: 53.85 | total time: 35.17m | eta: 75.9m step 01771/05568 (31.81%) | loss: 2.686092 | lrm: 1.00 | dt: 1195.38ms | tok/sec: 877,189 | bf16_mfu: 53.88 | total time: 35.19m | eta: 75.8m step 01772/05568 (31.82%) | loss: 2.687367 | lrm: 1.00 | dt: 1193.19ms | tok/sec: 878,801 | bf16_mfu: 53.98 | total time: 35.21m | eta: 75.8m step 01773/05568 (31.84%) | loss: 2.682544 | lrm: 1.00 | dt: 1190.74ms | tok/sec: 880,610 | bf16_mfu: 54.09 | total time: 35.23m | eta: 75.8m step 01774/05568 (31.86%) | loss: 2.687192 | lrm: 1.00 | dt: 1189.21ms | tok/sec: 881,739 | bf16_mfu: 54.16 | total time: 35.25m | eta: 75.8m step 01775/05568 (31.88%) | loss: 2.678749 | lrm: 1.00 | dt: 1195.35ms | tok/sec: 877,209 | bf16_mfu: 53.88 | total time: 35.27m | eta: 75.8m step 01776/05568 (31.90%) | loss: 2.681082 | lrm: 1.00 | dt: 1197.46ms | tok/sec: 875,670 | bf16_mfu: 53.79 | total time: 35.29m | eta: 75.7m step 01777/05568 (31.91%) | loss: 2.687355 | lrm: 1.00 | dt: 1194.51ms | tok/sec: 877,825 | bf16_mfu: 53.92 | total time: 35.31m | eta: 75.7m step 01778/05568 (31.93%) | loss: 2.689000 | lrm: 1.00 | dt: 1187.08ms | tok/sec: 883,321 | bf16_mfu: 54.26 | total time: 35.33m | eta: 75.7m step 01779/05568 (31.95%) | loss: 2.689540 | lrm: 1.00 | dt: 1199.78ms | tok/sec: 873,972 | bf16_mfu: 53.69 | total time: 35.35m | eta: 75.7m step 01780/05568 (31.97%) | loss: 2.687306 | lrm: 1.00 | dt: 1195.35ms | tok/sec: 877,212 | bf16_mfu: 53.89 | total time: 35.37m | eta: 75.7m step 01781/05568 (31.99%) | loss: 2.681514 | lrm: 1.00 | dt: 1195.89ms | tok/sec: 876,812 | bf16_mfu: 53.86 | total time: 35.39m | eta: 75.6m step 01782/05568 (32.00%) | loss: 2.685871 | lrm: 1.00 | dt: 1197.13ms | tok/sec: 875,905 | bf16_mfu: 53.80 | total time: 35.41m | eta: 75.6m step 01783/05568 (32.02%) | loss: 2.687166 | lrm: 1.00 | dt: 1193.89ms | tok/sec: 878,287 | bf16_mfu: 53.95 | total time: 35.43m | eta: 75.6m step 01784/05568 (32.04%) | loss: 2.688033 | lrm: 1.00 | dt: 1186.41ms | tok/sec: 883,823 | bf16_mfu: 54.29 | total time: 35.45m | eta: 75.6m step 01785/05568 (32.06%) | loss: 2.686299 | lrm: 1.00 | dt: 1198.95ms | tok/sec: 874,581 | bf16_mfu: 53.72 | total time: 35.47m | eta: 75.6m step 01786/05568 (32.08%) | loss: 2.682407 | lrm: 1.00 | dt: 1191.87ms | tok/sec: 879,772 | bf16_mfu: 54.04 | total time: 35.49m | eta: 75.5m step 01787/05568 (32.09%) | loss: 2.684945 | lrm: 1.00 | dt: 1191.24ms | tok/sec: 880,239 | bf16_mfu: 54.07 | total time: 35.51m | eta: 75.5m step 01788/05568 (32.11%) | loss: 2.686067 | lrm: 1.00 | dt: 1189.20ms | tok/sec: 881,745 | bf16_mfu: 54.16 | total time: 35.53m | eta: 75.5m step 01789/05568 (32.13%) | loss: 2.688379 | lrm: 1.00 | dt: 1191.35ms | tok/sec: 880,156 | bf16_mfu: 54.07 | total time: 35.55m | eta: 75.5m step 01790/05568 (32.15%) | loss: 2.693980 | lrm: 1.00 | dt: 1193.10ms | tok/sec: 878,863 | bf16_mfu: 53.99 | total time: 35.57m | eta: 75.5m step 01791/05568 (32.17%) | loss: 2.688845 | lrm: 1.00 | dt: 1193.87ms | tok/sec: 878,296 | bf16_mfu: 53.95 | total time: 35.58m | eta: 75.4m step 01792/05568 (32.18%) | loss: 2.677307 | lrm: 1.00 | dt: 1198.12ms | tok/sec: 875,181 | bf16_mfu: 53.76 | total time: 35.60m | eta: 75.4m step 01793/05568 (32.20%) | loss: 2.672205 | lrm: 1.00 | dt: 1189.92ms | tok/sec: 881,213 | bf16_mfu: 54.13 | total time: 35.62m | eta: 75.4m step 01794/05568 (32.22%) | loss: 2.668688 | lrm: 1.00 | dt: 1188.56ms | tok/sec: 882,221 | bf16_mfu: 54.19 | total time: 35.64m | eta: 75.4m step 01795/05568 (32.24%) | loss: 2.666764 | lrm: 1.00 | dt: 1190.05ms | tok/sec: 881,120 | bf16_mfu: 54.13 | total time: 35.66m | eta: 75.4m step 01796/05568 (32.26%) | loss: 2.659150 | lrm: 1.00 | dt: 1192.68ms | tok/sec: 879,175 | bf16_mfu: 54.01 | total time: 35.68m | eta: 75.3m step 01797/05568 (32.27%) | loss: 2.670015 | lrm: 1.00 | dt: 1190.35ms | tok/sec: 880,896 | bf16_mfu: 54.11 | total time: 35.70m | eta: 75.3m step 01798/05568 (32.29%) | loss: 2.671404 | lrm: 1.00 | dt: 1190.04ms | tok/sec: 881,123 | bf16_mfu: 54.13 | total time: 35.72m | eta: 75.3m step 01799/05568 (32.31%) | loss: 2.672612 | lrm: 1.00 | dt: 1189.92ms | tok/sec: 881,212 | bf16_mfu: 54.13 | total time: 35.74m | eta: 75.3m step 01800/05568 (32.33%) | loss: 2.677031 | lrm: 1.00 | dt: 1193.73ms | tok/sec: 878,405 | bf16_mfu: 53.96 | total time: 35.76m | eta: 75.3m step 01801/05568 (32.35%) | loss: 2.679060 | lrm: 1.00 | dt: 1197.62ms | tok/sec: 875,549 | bf16_mfu: 53.78 | total time: 35.78m | eta: 75.2m step 01802/05568 (32.36%) | loss: 2.685674 | lrm: 1.00 | dt: 1196.79ms | tok/sec: 876,154 | bf16_mfu: 53.82 | total time: 35.80m | eta: 75.2m step 01803/05568 (32.38%) | loss: 2.670285 | lrm: 1.00 | dt: 1193.92ms | tok/sec: 878,263 | bf16_mfu: 53.95 | total time: 35.82m | eta: 75.2m step 01804/05568 (32.40%) | loss: 2.666504 | lrm: 1.00 | dt: 1195.96ms | tok/sec: 876,763 | bf16_mfu: 53.86 | total time: 35.84m | eta: 75.2m step 01805/05568 (32.42%) | loss: 2.660037 | lrm: 1.00 | dt: 1187.47ms | tok/sec: 883,030 | bf16_mfu: 54.24 | total time: 35.86m | eta: 75.2m step 01806/05568 (32.44%) | loss: 2.662324 | lrm: 1.00 | dt: 1198.89ms | tok/sec: 874,622 | bf16_mfu: 53.73 | total time: 35.88m | eta: 75.1m step 01807/05568 (32.45%) | loss: 2.661864 | lrm: 1.00 | dt: 1195.67ms | tok/sec: 876,981 | bf16_mfu: 53.87 | total time: 35.90m | eta: 75.1m step 01808/05568 (32.47%) | loss: 2.661807 | lrm: 1.00 | dt: 1190.94ms | tok/sec: 880,457 | bf16_mfu: 54.08 | total time: 35.92m | eta: 75.1m step 01809/05568 (32.49%) | loss: 2.666005 | lrm: 1.00 | dt: 1189.61ms | tok/sec: 881,444 | bf16_mfu: 54.15 | total time: 35.94m | eta: 75.1m step 01810/05568 (32.51%) | loss: 2.679690 | lrm: 1.00 | dt: 1195.88ms | tok/sec: 876,824 | bf16_mfu: 53.86 | total time: 35.96m | eta: 75.1m step 01811/05568 (32.53%) | loss: 2.676055 | lrm: 1.00 | dt: 1189.93ms | tok/sec: 881,206 | bf16_mfu: 54.13 | total time: 35.98m | eta: 75.0m step 01812/05568 (32.54%) | loss: 2.672467 | lrm: 1.00 | dt: 1188.66ms | tok/sec: 882,148 | bf16_mfu: 54.19 | total time: 36.00m | eta: 75.0m step 01813/05568 (32.56%) | loss: 2.679744 | lrm: 1.00 | dt: 1200.21ms | tok/sec: 873,662 | bf16_mfu: 53.67 | total time: 36.02m | eta: 75.0m step 01814/05568 (32.58%) | loss: 2.682004 | lrm: 1.00 | dt: 1193.63ms | tok/sec: 878,479 | bf16_mfu: 53.96 | total time: 36.04m | eta: 75.0m step 01815/05568 (32.60%) | loss: 2.680681 | lrm: 1.00 | dt: 1191.41ms | tok/sec: 880,114 | bf16_mfu: 54.06 | total time: 36.06m | eta: 75.0m step 01816/05568 (32.61%) | loss: 2.672339 | lrm: 1.00 | dt: 1194.19ms | tok/sec: 878,061 | bf16_mfu: 53.94 | total time: 36.08m | eta: 74.9m step 01817/05568 (32.63%) | loss: 2.675463 | lrm: 1.00 | dt: 1195.85ms | tok/sec: 876,843 | bf16_mfu: 53.86 | total time: 36.10m | eta: 74.9m step 01818/05568 (32.65%) | loss: 2.668903 | lrm: 1.00 | dt: 1191.63ms | tok/sec: 879,952 | bf16_mfu: 54.05 | total time: 36.12m | eta: 74.9m step 01819/05568 (32.67%) | loss: 2.669026 | lrm: 1.00 | dt: 1195.10ms | tok/sec: 877,392 | bf16_mfu: 53.90 | total time: 36.14m | eta: 74.9m step 01820/05568 (32.69%) | loss: 2.660174 | lrm: 1.00 | dt: 1198.89ms | tok/sec: 874,618 | bf16_mfu: 53.73 | total time: 36.16m | eta: 74.9m step 01821/05568 (32.70%) | loss: 2.671714 | lrm: 1.00 | dt: 1196.97ms | tok/sec: 876,026 | bf16_mfu: 53.81 | total time: 36.18m | eta: 74.8m step 01822/05568 (32.72%) | loss: 2.676594 | lrm: 1.00 | dt: 1190.58ms | tok/sec: 880,723 | bf16_mfu: 54.10 | total time: 36.20m | eta: 74.8m step 01823/05568 (32.74%) | loss: 2.675992 | lrm: 1.00 | dt: 1189.92ms | tok/sec: 881,215 | bf16_mfu: 54.13 | total time: 36.22m | eta: 74.8m step 01824/05568 (32.76%) | loss: 2.672576 | lrm: 1.00 | dt: 1200.14ms | tok/sec: 873,710 | bf16_mfu: 53.67 | total time: 36.24m | eta: 74.8m step 01825/05568 (32.78%) | loss: 2.670826 | lrm: 1.00 | dt: 1192.37ms | tok/sec: 879,408 | bf16_mfu: 54.02 | total time: 36.26m | eta: 74.8m step 01826/05568 (32.79%) | loss: 2.667748 | lrm: 1.00 | dt: 1195.99ms | tok/sec: 876,743 | bf16_mfu: 53.86 | total time: 36.28m | eta: 74.7m step 01827/05568 (32.81%) | loss: 2.667933 | lrm: 1.00 | dt: 1191.19ms | tok/sec: 880,273 | bf16_mfu: 54.07 | total time: 36.30m | eta: 74.7m step 01828/05568 (32.83%) | loss: 2.684988 | lrm: 1.00 | dt: 1192.14ms | tok/sec: 879,578 | bf16_mfu: 54.03 | total time: 36.32m | eta: 74.7m step 01829/05568 (32.85%) | loss: 2.693350 | lrm: 1.00 | dt: 1192.03ms | tok/sec: 879,656 | bf16_mfu: 54.04 | total time: 36.34m | eta: 74.7m step 01830/05568 (32.87%) | loss: 2.686234 | lrm: 1.00 | dt: 1194.70ms | tok/sec: 877,687 | bf16_mfu: 53.91 | total time: 36.36m | eta: 74.7m step 01831/05568 (32.88%) | loss: 2.686606 | lrm: 1.00 | dt: 1191.54ms | tok/sec: 880,019 | bf16_mfu: 54.06 | total time: 36.38m | eta: 74.6m step 01832/05568 (32.90%) | loss: 2.683433 | lrm: 1.00 | dt: 1197.66ms | tok/sec: 875,521 | bf16_mfu: 53.78 | total time: 36.40m | eta: 74.6m step 01833/05568 (32.92%) | loss: 2.692746 | lrm: 1.00 | dt: 1188.24ms | tok/sec: 882,464 | bf16_mfu: 54.21 | total time: 36.42m | eta: 74.6m step 01834/05568 (32.94%) | loss: 2.694921 | lrm: 1.00 | dt: 1196.17ms | tok/sec: 876,608 | bf16_mfu: 53.85 | total time: 36.44m | eta: 74.6m step 01835/05568 (32.96%) | loss: 2.686053 | lrm: 1.00 | dt: 1199.62ms | tok/sec: 874,092 | bf16_mfu: 53.69 | total time: 36.46m | eta: 74.6m step 01836/05568 (32.97%) | loss: 2.688361 | lrm: 1.00 | dt: 1190.62ms | tok/sec: 880,698 | bf16_mfu: 54.10 | total time: 36.48m | eta: 74.5m step 01837/05568 (32.99%) | loss: 2.674513 | lrm: 1.00 | dt: 1191.78ms | tok/sec: 879,837 | bf16_mfu: 54.05 | total time: 36.50m | eta: 74.5m step 01838/05568 (33.01%) | loss: 2.683632 | lrm: 1.00 | dt: 1194.12ms | tok/sec: 878,118 | bf16_mfu: 53.94 | total time: 36.52m | eta: 74.5m step 01839/05568 (33.03%) | loss: 2.686503 | lrm: 1.00 | dt: 1192.80ms | tok/sec: 879,087 | bf16_mfu: 54.00 | total time: 36.54m | eta: 74.5m step 01840/05568 (33.05%) | loss: 2.681285 | lrm: 1.00 | dt: 1189.87ms | tok/sec: 881,254 | bf16_mfu: 54.13 | total time: 36.56m | eta: 74.5m step 01841/05568 (33.06%) | loss: 2.679741 | lrm: 1.00 | dt: 1196.07ms | tok/sec: 876,682 | bf16_mfu: 53.85 | total time: 36.58m | eta: 74.4m step 01842/05568 (33.08%) | loss: 2.665978 | lrm: 1.00 | dt: 1189.38ms | tok/sec: 881,618 | bf16_mfu: 54.16 | total time: 36.60m | eta: 74.4m step 01843/05568 (33.10%) | loss: 2.667944 | lrm: 1.00 | dt: 1190.50ms | tok/sec: 880,789 | bf16_mfu: 54.10 | total time: 36.62m | eta: 74.4m step 01844/05568 (33.12%) | loss: 2.661517 | lrm: 1.00 | dt: 1192.15ms | tok/sec: 879,570 | bf16_mfu: 54.03 | total time: 36.64m | eta: 74.4m step 01845/05568 (33.14%) | loss: 2.669482 | lrm: 1.00 | dt: 1195.37ms | tok/sec: 877,196 | bf16_mfu: 53.88 | total time: 36.66m | eta: 74.4m step 01846/05568 (33.15%) | loss: 2.677141 | lrm: 1.00 | dt: 1196.43ms | tok/sec: 876,419 | bf16_mfu: 53.84 | total time: 36.68m | eta: 74.3m step 01847/05568 (33.17%) | loss: 2.671295 | lrm: 1.00 | dt: 1193.59ms | tok/sec: 878,504 | bf16_mfu: 53.96 | total time: 36.70m | eta: 74.3m step 01848/05568 (33.19%) | loss: 2.671535 | lrm: 1.00 | dt: 1194.77ms | tok/sec: 877,637 | bf16_mfu: 53.91 | total time: 36.72m | eta: 74.3m step 01849/05568 (33.21%) | loss: 2.669746 | lrm: 1.00 | dt: 1192.33ms | tok/sec: 879,433 | bf16_mfu: 54.02 | total time: 36.74m | eta: 74.3m step 01850/05568 (33.23%) | loss: 2.664401 | lrm: 1.00 | dt: 1194.26ms | tok/sec: 878,016 | bf16_mfu: 53.93 | total time: 36.76m | eta: 74.3m step 01851/05568 (33.24%) | loss: 2.671827 | lrm: 1.00 | dt: 1193.06ms | tok/sec: 878,898 | bf16_mfu: 53.99 | total time: 36.78m | eta: 74.2m step 01852/05568 (33.26%) | loss: 2.666190 | lrm: 1.00 | dt: 1194.20ms | tok/sec: 878,053 | bf16_mfu: 53.94 | total time: 36.80m | eta: 74.2m step 01853/05568 (33.28%) | loss: 2.665511 | lrm: 1.00 | dt: 1195.14ms | tok/sec: 877,364 | bf16_mfu: 53.89 | total time: 36.82m | eta: 74.2m step 01854/05568 (33.30%) | loss: 2.665406 | lrm: 1.00 | dt: 1192.67ms | tok/sec: 879,185 | bf16_mfu: 54.01 | total time: 36.84m | eta: 74.2m step 01855/05568 (33.32%) | loss: 2.669888 | lrm: 1.00 | dt: 1192.12ms | tok/sec: 879,587 | bf16_mfu: 54.03 | total time: 36.86m | eta: 74.2m step 01856/05568 (33.33%) | loss: 2.666067 | lrm: 1.00 | dt: 1191.78ms | tok/sec: 879,839 | bf16_mfu: 54.05 | total time: 36.88m | eta: 74.1m step 01857/05568 (33.35%) | loss: 2.665688 | lrm: 1.00 | dt: 1194.03ms | tok/sec: 878,181 | bf16_mfu: 53.94 | total time: 36.90m | eta: 74.1m step 01858/05568 (33.37%) | loss: 2.668281 | lrm: 1.00 | dt: 1191.95ms | tok/sec: 879,714 | bf16_mfu: 54.04 | total time: 36.92m | eta: 74.1m step 01859/05568 (33.39%) | loss: 2.665347 | lrm: 1.00 | dt: 1192.57ms | tok/sec: 879,254 | bf16_mfu: 54.01 | total time: 36.94m | eta: 74.1m step 01860/05568 (33.41%) | loss: 2.673874 | lrm: 1.00 | dt: 1195.61ms | tok/sec: 877,025 | bf16_mfu: 53.87 | total time: 36.96m | eta: 74.1m step 01861/05568 (33.42%) | loss: 2.673671 | lrm: 1.00 | dt: 1189.37ms | tok/sec: 881,623 | bf16_mfu: 54.16 | total time: 36.98m | eta: 74.0m step 01862/05568 (33.44%) | loss: 2.679444 | lrm: 1.00 | dt: 1201.38ms | tok/sec: 872,812 | bf16_mfu: 53.61 | total time: 37.00m | eta: 74.0m step 01863/05568 (33.46%) | loss: 2.668909 | lrm: 1.00 | dt: 1190.20ms | tok/sec: 881,008 | bf16_mfu: 54.12 | total time: 37.02m | eta: 74.0m step 01864/05568 (33.48%) | loss: 2.668254 | lrm: 1.00 | dt: 1191.46ms | tok/sec: 880,076 | bf16_mfu: 54.06 | total time: 37.04m | eta: 74.0m step 01865/05568 (33.49%) | loss: 2.654436 | lrm: 1.00 | dt: 1199.02ms | tok/sec: 874,524 | bf16_mfu: 53.72 | total time: 37.06m | eta: 74.0m step 01866/05568 (33.51%) | loss: 2.654728 | lrm: 1.00 | dt: 1192.96ms | tok/sec: 878,970 | bf16_mfu: 53.99 | total time: 37.08m | eta: 73.9m step 01867/05568 (33.53%) | loss: 2.646022 | lrm: 1.00 | dt: 1190.68ms | tok/sec: 880,650 | bf16_mfu: 54.10 | total time: 37.10m | eta: 73.9m step 01868/05568 (33.55%) | loss: 2.643030 | lrm: 1.00 | dt: 1190.69ms | tok/sec: 880,648 | bf16_mfu: 54.10 | total time: 37.12m | eta: 73.9m step 01869/05568 (33.57%) | loss: 2.645841 | lrm: 1.00 | dt: 1188.68ms | tok/sec: 882,136 | bf16_mfu: 54.19 | total time: 37.14m | eta: 73.9m step 01870/05568 (33.58%) | loss: 2.637906 | lrm: 1.00 | dt: 1189.71ms | tok/sec: 881,371 | bf16_mfu: 54.14 | total time: 37.16m | eta: 73.9m step 01871/05568 (33.60%) | loss: 2.624189 | lrm: 1.00 | dt: 1190.83ms | tok/sec: 880,545 | bf16_mfu: 54.09 | total time: 37.18m | eta: 73.8m step 01872/05568 (33.62%) | loss: 2.626355 | lrm: 1.00 | dt: 1195.23ms | tok/sec: 877,303 | bf16_mfu: 53.89 | total time: 37.20m | eta: 73.8m step 01873/05568 (33.64%) | loss: 2.626451 | lrm: 1.00 | dt: 1190.99ms | tok/sec: 880,426 | bf16_mfu: 54.08 | total time: 37.22m | eta: 73.8m step 01874/05568 (33.66%) | loss: 2.634235 | lrm: 1.00 | dt: 1193.64ms | tok/sec: 878,472 | bf16_mfu: 53.96 | total time: 37.24m | eta: 73.8m step 01875/05568 (33.67%) | loss: 2.643180 | lrm: 1.00 | dt: 1191.59ms | tok/sec: 879,978 | bf16_mfu: 54.06 | total time: 37.26m | eta: 73.8m step 01876/05568 (33.69%) | loss: 2.644944 | lrm: 1.00 | dt: 1191.21ms | tok/sec: 880,258 | bf16_mfu: 54.07 | total time: 37.28m | eta: 73.7m step 01877/05568 (33.71%) | loss: 2.642156 | lrm: 1.00 | dt: 1196.40ms | tok/sec: 876,443 | bf16_mfu: 53.84 | total time: 37.30m | eta: 73.7m step 01878/05568 (33.73%) | loss: 2.647214 | lrm: 1.00 | dt: 1194.37ms | tok/sec: 877,934 | bf16_mfu: 53.93 | total time: 37.32m | eta: 73.7m step 01879/05568 (33.75%) | loss: 2.643446 | lrm: 1.00 | dt: 1194.53ms | tok/sec: 877,815 | bf16_mfu: 53.92 | total time: 37.34m | eta: 73.7m step 01880/05568 (33.76%) | loss: 2.644354 | lrm: 1.00 | dt: 1194.34ms | tok/sec: 877,957 | bf16_mfu: 53.93 | total time: 37.35m | eta: 73.7m step 01881/05568 (33.78%) | loss: 2.640006 | lrm: 1.00 | dt: 1189.86ms | tok/sec: 881,263 | bf16_mfu: 54.13 | total time: 37.37m | eta: 73.6m step 01882/05568 (33.80%) | loss: 2.645312 | lrm: 1.00 | dt: 1195.76ms | tok/sec: 876,909 | bf16_mfu: 53.87 | total time: 37.39m | eta: 73.6m step 01883/05568 (33.82%) | loss: 2.641579 | lrm: 1.00 | dt: 1197.40ms | tok/sec: 875,710 | bf16_mfu: 53.79 | total time: 37.41m | eta: 73.6m step 01884/05568 (33.84%) | loss: 2.653493 | lrm: 1.00 | dt: 1189.33ms | tok/sec: 881,656 | bf16_mfu: 54.16 | total time: 37.43m | eta: 73.6m step 01885/05568 (33.85%) | loss: 2.665730 | lrm: 1.00 | dt: 1196.91ms | tok/sec: 876,068 | bf16_mfu: 53.81 | total time: 37.45m | eta: 73.6m step 01886/05568 (33.87%) | loss: 2.664114 | lrm: 1.00 | dt: 1191.50ms | tok/sec: 880,049 | bf16_mfu: 54.06 | total time: 37.47m | eta: 73.5m step 01887/05568 (33.89%) | loss: 2.656761 | lrm: 1.00 | dt: 1191.28ms | tok/sec: 880,208 | bf16_mfu: 54.07 | total time: 37.49m | eta: 73.5m step 01888/05568 (33.91%) | loss: 2.657264 | lrm: 1.00 | dt: 1199.13ms | tok/sec: 874,446 | bf16_mfu: 53.72 | total time: 37.51m | eta: 73.5m step 01889/05568 (33.93%) | loss: 2.661819 | lrm: 1.00 | dt: 1198.16ms | tok/sec: 875,156 | bf16_mfu: 53.76 | total time: 37.53m | eta: 73.5m step 01890/05568 (33.94%) | loss: 2.664018 | lrm: 1.00 | dt: 1196.03ms | tok/sec: 876,713 | bf16_mfu: 53.85 | total time: 37.55m | eta: 73.5m step 01891/05568 (33.96%) | loss: 2.664540 | lrm: 1.00 | dt: 1189.13ms | tok/sec: 881,802 | bf16_mfu: 54.17 | total time: 37.57m | eta: 73.4m step 01892/05568 (33.98%) | loss: 2.666927 | lrm: 1.00 | dt: 1195.41ms | tok/sec: 877,166 | bf16_mfu: 53.88 | total time: 37.59m | eta: 73.4m step 01893/05568 (34.00%) | loss: 2.677018 | lrm: 1.00 | dt: 1198.33ms | tok/sec: 875,028 | bf16_mfu: 53.75 | total time: 37.61m | eta: 73.4m step 01894/05568 (34.02%) | loss: 2.678819 | lrm: 1.00 | dt: 1198.42ms | tok/sec: 874,964 | bf16_mfu: 53.75 | total time: 37.63m | eta: 73.4m step 01895/05568 (34.03%) | loss: 2.677001 | lrm: 1.00 | dt: 1186.76ms | tok/sec: 883,561 | bf16_mfu: 54.28 | total time: 37.65m | eta: 73.3m step 01896/05568 (34.05%) | loss: 2.685014 | lrm: 1.00 | dt: 1191.44ms | tok/sec: 880,092 | bf16_mfu: 54.06 | total time: 37.67m | eta: 73.3m step 01897/05568 (34.07%) | loss: 2.681705 | lrm: 1.00 | dt: 1193.22ms | tok/sec: 878,778 | bf16_mfu: 53.98 | total time: 37.69m | eta: 73.3m step 01898/05568 (34.09%) | loss: 2.684584 | lrm: 1.00 | dt: 1191.96ms | tok/sec: 879,708 | bf16_mfu: 54.04 | total time: 37.71m | eta: 73.3m step 01899/05568 (34.11%) | loss: 2.690926 | lrm: 1.00 | dt: 1191.66ms | tok/sec: 879,927 | bf16_mfu: 54.05 | total time: 37.73m | eta: 73.3m step 01900/05568 (34.12%) | loss: 2.690947 | lrm: 1.00 | dt: 1195.92ms | tok/sec: 876,792 | bf16_mfu: 53.86 | total time: 37.75m | eta: 73.2m step 01901/05568 (34.14%) | loss: 2.689452 | lrm: 1.00 | dt: 1195.49ms | tok/sec: 877,108 | bf16_mfu: 53.88 | total time: 37.77m | eta: 73.2m step 01902/05568 (34.16%) | loss: 2.683577 | lrm: 1.00 | dt: 1197.02ms | tok/sec: 875,989 | bf16_mfu: 53.81 | total time: 37.79m | eta: 73.2m step 01903/05568 (34.18%) | loss: 2.684653 | lrm: 1.00 | dt: 1191.96ms | tok/sec: 879,705 | bf16_mfu: 54.04 | total time: 37.81m | eta: 73.2m step 01904/05568 (34.20%) | loss: 2.681750 | lrm: 1.00 | dt: 1193.86ms | tok/sec: 878,305 | bf16_mfu: 53.95 | total time: 37.83m | eta: 73.2m step 01905/05568 (34.21%) | loss: 2.677066 | lrm: 1.00 | dt: 1196.50ms | tok/sec: 876,372 | bf16_mfu: 53.83 | total time: 37.85m | eta: 73.1m step 01906/05568 (34.23%) | loss: 2.674944 | lrm: 1.00 | dt: 1192.46ms | tok/sec: 879,337 | bf16_mfu: 54.02 | total time: 37.87m | eta: 73.1m step 01907/05568 (34.25%) | loss: 2.678417 | lrm: 1.00 | dt: 1192.71ms | tok/sec: 879,153 | bf16_mfu: 54.00 | total time: 37.89m | eta: 73.1m step 01908/05568 (34.27%) | loss: 2.680653 | lrm: 1.00 | dt: 1191.72ms | tok/sec: 879,886 | bf16_mfu: 54.05 | total time: 37.91m | eta: 73.1m step 01909/05568 (34.29%) | loss: 2.682723 | lrm: 1.00 | dt: 1192.96ms | tok/sec: 878,970 | bf16_mfu: 53.99 | total time: 37.93m | eta: 73.1m step 01910/05568 (34.30%) | loss: 2.685614 | lrm: 1.00 | dt: 1191.33ms | tok/sec: 880,169 | bf16_mfu: 54.07 | total time: 37.95m | eta: 73.0m step 01911/05568 (34.32%) | loss: 2.678456 | lrm: 1.00 | dt: 1195.30ms | tok/sec: 877,250 | bf16_mfu: 53.89 | total time: 37.97m | eta: 73.0m step 01912/05568 (34.34%) | loss: 2.676659 | lrm: 1.00 | dt: 1192.88ms | tok/sec: 879,028 | bf16_mfu: 54.00 | total time: 37.99m | eta: 73.0m step 01913/05568 (34.36%) | loss: 2.683838 | lrm: 1.00 | dt: 1191.06ms | tok/sec: 880,369 | bf16_mfu: 54.08 | total time: 38.01m | eta: 73.0m step 01914/05568 (34.38%) | loss: 2.684994 | lrm: 1.00 | dt: 1193.62ms | tok/sec: 878,485 | bf16_mfu: 53.96 | total time: 38.03m | eta: 73.0m step 01915/05568 (34.39%) | loss: 2.669249 | lrm: 1.00 | dt: 1195.31ms | tok/sec: 877,243 | bf16_mfu: 53.89 | total time: 38.05m | eta: 72.9m step 01916/05568 (34.41%) | loss: 2.676912 | lrm: 1.00 | dt: 1193.17ms | tok/sec: 878,813 | bf16_mfu: 53.98 | total time: 38.07m | eta: 72.9m step 01917/05568 (34.43%) | loss: 2.664549 | lrm: 1.00 | dt: 1191.89ms | tok/sec: 879,758 | bf16_mfu: 54.04 | total time: 38.09m | eta: 72.9m step 01918/05568 (34.45%) | loss: 2.665907 | lrm: 1.00 | dt: 1196.38ms | tok/sec: 876,459 | bf16_mfu: 53.84 | total time: 38.11m | eta: 72.9m step 01919/05568 (34.46%) | loss: 2.664192 | lrm: 1.00 | dt: 1193.49ms | tok/sec: 878,576 | bf16_mfu: 53.97 | total time: 38.13m | eta: 72.9m step 01920/05568 (34.48%) | loss: 2.656109 | lrm: 1.00 | dt: 1191.08ms | tok/sec: 880,354 | bf16_mfu: 54.08 | total time: 38.15m | eta: 72.8m step 01921/05568 (34.50%) | loss: 2.656427 | lrm: 1.00 | dt: 1197.91ms | tok/sec: 875,336 | bf16_mfu: 53.77 | total time: 38.17m | eta: 72.8m step 01922/05568 (34.52%) | loss: 2.651143 | lrm: 1.00 | dt: 1197.42ms | tok/sec: 875,695 | bf16_mfu: 53.79 | total time: 38.19m | eta: 72.8m step 01923/05568 (34.54%) | loss: 2.645610 | lrm: 1.00 | dt: 1196.76ms | tok/sec: 876,175 | bf16_mfu: 53.82 | total time: 38.21m | eta: 72.8m step 01924/05568 (34.55%) | loss: 2.649438 | lrm: 1.00 | dt: 1191.71ms | tok/sec: 879,892 | bf16_mfu: 54.05 | total time: 38.23m | eta: 72.8m step 01925/05568 (34.57%) | loss: 2.649633 | lrm: 1.00 | dt: 1193.38ms | tok/sec: 878,659 | bf16_mfu: 53.97 | total time: 38.25m | eta: 72.7m step 01926/05568 (34.59%) | loss: 2.652873 | lrm: 1.00 | dt: 1194.34ms | tok/sec: 877,952 | bf16_mfu: 53.93 | total time: 38.27m | eta: 72.7m step 01927/05568 (34.61%) | loss: 2.653717 | lrm: 1.00 | dt: 1192.68ms | tok/sec: 879,173 | bf16_mfu: 54.01 | total time: 38.29m | eta: 72.7m step 01928/05568 (34.63%) | loss: 2.642537 | lrm: 1.00 | dt: 1189.53ms | tok/sec: 881,503 | bf16_mfu: 54.15 | total time: 38.31m | eta: 72.7m step 01929/05568 (34.64%) | loss: 2.647324 | lrm: 1.00 | dt: 1195.28ms | tok/sec: 877,263 | bf16_mfu: 53.89 | total time: 38.33m | eta: 72.7m step 01930/05568 (34.66%) | loss: 2.645575 | lrm: 1.00 | dt: 1186.11ms | tok/sec: 884,044 | bf16_mfu: 54.30 | total time: 38.35m | eta: 72.6m step 01931/05568 (34.68%) | loss: 2.642497 | lrm: 1.00 | dt: 1195.27ms | tok/sec: 877,269 | bf16_mfu: 53.89 | total time: 38.37m | eta: 72.6m step 01932/05568 (34.70%) | loss: 2.642870 | lrm: 1.00 | dt: 1194.59ms | tok/sec: 877,770 | bf16_mfu: 53.92 | total time: 38.39m | eta: 72.6m step 01933/05568 (34.72%) | loss: 2.651447 | lrm: 1.00 | dt: 1191.34ms | tok/sec: 880,165 | bf16_mfu: 54.07 | total time: 38.41m | eta: 72.6m step 01934/05568 (34.73%) | loss: 2.643746 | lrm: 1.00 | dt: 1195.05ms | tok/sec: 877,435 | bf16_mfu: 53.90 | total time: 38.43m | eta: 72.6m step 01935/05568 (34.75%) | loss: 2.639682 | lrm: 1.00 | dt: 1186.13ms | tok/sec: 884,034 | bf16_mfu: 54.30 | total time: 38.45m | eta: 72.5m step 01936/05568 (34.77%) | loss: 2.644311 | lrm: 1.00 | dt: 1195.49ms | tok/sec: 877,106 | bf16_mfu: 53.88 | total time: 38.47m | eta: 72.5m step 01937/05568 (34.79%) | loss: 2.650281 | lrm: 1.00 | dt: 1194.03ms | tok/sec: 878,183 | bf16_mfu: 53.94 | total time: 38.49m | eta: 72.5m step 01938/05568 (34.81%) | loss: 2.657919 | lrm: 1.00 | dt: 1190.40ms | tok/sec: 880,861 | bf16_mfu: 54.11 | total time: 38.51m | eta: 72.5m step 01939/05568 (34.82%) | loss: 2.664585 | lrm: 1.00 | dt: 1196.42ms | tok/sec: 876,424 | bf16_mfu: 53.84 | total time: 38.53m | eta: 72.5m step 01940/05568 (34.84%) | loss: 2.655160 | lrm: 1.00 | dt: 1199.53ms | tok/sec: 874,152 | bf16_mfu: 53.70 | total time: 38.55m | eta: 72.4m step 01941/05568 (34.86%) | loss: 2.650838 | lrm: 1.00 | dt: 1188.98ms | tok/sec: 881,909 | bf16_mfu: 54.17 | total time: 38.57m | eta: 72.4m step 01942/05568 (34.88%) | loss: 2.644602 | lrm: 1.00 | dt: 1199.20ms | tok/sec: 874,398 | bf16_mfu: 53.71 | total time: 38.59m | eta: 72.4m step 01943/05568 (34.90%) | loss: 2.641640 | lrm: 1.00 | dt: 1192.12ms | tok/sec: 879,585 | bf16_mfu: 54.03 | total time: 38.61m | eta: 72.4m step 01944/05568 (34.91%) | loss: 2.639240 | lrm: 1.00 | dt: 1194.28ms | tok/sec: 878,000 | bf16_mfu: 53.93 | total time: 38.63m | eta: 72.4m step 01945/05568 (34.93%) | loss: 2.630474 | lrm: 1.00 | dt: 1193.53ms | tok/sec: 878,548 | bf16_mfu: 53.97 | total time: 38.65m | eta: 72.3m step 01946/05568 (34.95%) | loss: 2.632892 | lrm: 1.00 | dt: 1191.60ms | tok/sec: 879,973 | bf16_mfu: 54.05 | total time: 38.67m | eta: 72.3m step 01947/05568 (34.97%) | loss: 2.642553 | lrm: 1.00 | dt: 1195.11ms | tok/sec: 877,392 | bf16_mfu: 53.90 | total time: 38.69m | eta: 72.3m step 01948/05568 (34.99%) | loss: 2.644658 | lrm: 1.00 | dt: 1193.85ms | tok/sec: 878,318 | bf16_mfu: 53.95 | total time: 38.71m | eta: 72.3m step 01949/05568 (35.00%) | loss: 2.648384 | lrm: 1.00 | dt: 1192.92ms | tok/sec: 878,997 | bf16_mfu: 53.99 | total time: 38.73m | eta: 72.3m checkpoint captured at step 1950 (17.0s) step 01950/05568 (35.02%) | loss: 2.661988 | lrm: 1.00 | dt: 1153.90ms | tok/sec: 908,726 | bf16_mfu: 55.82 | total time: 38.75m | eta: 72.2m step 01951/05568 (35.04%) | loss: 2.659208 | lrm: 1.00 | dt: 1175.07ms | tok/sec: 892,355 | bf16_mfu: 54.82 | total time: 38.77m | eta: 72.2m step 01952/05568 (35.06%) | loss: 2.663603 | lrm: 1.00 | dt: 1187.55ms | tok/sec: 882,976 | bf16_mfu: 54.24 | total time: 38.79m | eta: 72.2m step 01953/05568 (35.08%) | loss: 2.682244 | lrm: 1.00 | dt: 1187.75ms | tok/sec: 882,823 | bf16_mfu: 54.23 | total time: 38.81m | eta: 72.2m step 01954/05568 (35.09%) | loss: 2.685417 | lrm: 1.00 | dt: 1192.46ms | tok/sec: 879,341 | bf16_mfu: 54.02 | total time: 38.83m | eta: 72.2m step 01955/05568 (35.11%) | loss: 2.676934 | lrm: 1.00 | dt: 1190.23ms | tok/sec: 880,982 | bf16_mfu: 54.12 | total time: 38.85m | eta: 72.1m step 01956/05568 (35.13%) | loss: 2.676866 | lrm: 1.00 | dt: 1187.34ms | tok/sec: 883,129 | bf16_mfu: 54.25 | total time: 38.87m | eta: 72.1m step 01957/05568 (35.15%) | loss: 2.671799 | lrm: 1.00 | dt: 1189.16ms | tok/sec: 881,780 | bf16_mfu: 54.17 | total time: 38.89m | eta: 72.1m step 01958/05568 (35.17%) | loss: 2.673814 | lrm: 1.00 | dt: 1192.41ms | tok/sec: 879,376 | bf16_mfu: 54.02 | total time: 38.91m | eta: 72.1m step 01959/05568 (35.18%) | loss: 2.667035 | lrm: 1.00 | dt: 1192.73ms | tok/sec: 879,141 | bf16_mfu: 54.00 | total time: 38.93m | eta: 72.1m step 01960/05568 (35.20%) | loss: 2.678399 | lrm: 1.00 | dt: 1190.77ms | tok/sec: 880,584 | bf16_mfu: 54.09 | total time: 38.94m | eta: 72.0m step 01961/05568 (35.22%) | loss: 2.665005 | lrm: 1.00 | dt: 1191.58ms | tok/sec: 879,986 | bf16_mfu: 54.06 | total time: 38.96m | eta: 72.0m step 01962/05568 (35.24%) | loss: 2.661254 | lrm: 1.00 | dt: 1188.95ms | tok/sec: 881,935 | bf16_mfu: 54.18 | total time: 38.98m | eta: 72.0m step 01963/05568 (35.26%) | loss: 2.662942 | lrm: 1.00 | dt: 1188.91ms | tok/sec: 881,966 | bf16_mfu: 54.18 | total time: 39.00m | eta: 72.0m step 01964/05568 (35.27%) | loss: 2.662939 | lrm: 1.00 | dt: 1185.22ms | tok/sec: 884,711 | bf16_mfu: 54.35 | total time: 39.02m | eta: 72.0m step 01965/05568 (35.29%) | loss: 2.668705 | lrm: 1.00 | dt: 1191.61ms | tok/sec: 879,967 | bf16_mfu: 54.05 | total time: 39.04m | eta: 71.9m step 01966/05568 (35.31%) | loss: 2.662334 | lrm: 1.00 | dt: 1188.45ms | tok/sec: 882,302 | bf16_mfu: 54.20 | total time: 39.06m | eta: 71.9m step 01967/05568 (35.33%) | loss: 2.667797 | lrm: 1.00 | dt: 1186.65ms | tok/sec: 883,644 | bf16_mfu: 54.28 | total time: 39.08m | eta: 71.9m step 01968/05568 (35.34%) | loss: 2.654878 | lrm: 1.00 | dt: 1188.69ms | tok/sec: 882,126 | bf16_mfu: 54.19 | total time: 39.10m | eta: 71.9m step 01969/05568 (35.36%) | loss: 2.665728 | lrm: 0.99 | dt: 1187.64ms | tok/sec: 882,910 | bf16_mfu: 54.24 | total time: 39.12m | eta: 71.9m step 01970/05568 (35.38%) | loss: 2.666660 | lrm: 0.99 | dt: 1191.78ms | tok/sec: 879,842 | bf16_mfu: 54.05 | total time: 39.14m | eta: 71.8m step 01971/05568 (35.40%) | loss: 2.653598 | lrm: 0.99 | dt: 1192.77ms | tok/sec: 879,106 | bf16_mfu: 54.00 | total time: 39.16m | eta: 71.8m step 01972/05568 (35.42%) | loss: 2.654241 | lrm: 0.99 | dt: 1186.53ms | tok/sec: 883,730 | bf16_mfu: 54.29 | total time: 39.18m | eta: 71.8m step 01973/05568 (35.43%) | loss: 2.658160 | lrm: 0.99 | dt: 1193.41ms | tok/sec: 878,636 | bf16_mfu: 53.97 | total time: 39.20m | eta: 71.8m step 01974/05568 (35.45%) | loss: 2.660868 | lrm: 0.99 | dt: 1195.93ms | tok/sec: 876,784 | bf16_mfu: 53.86 | total time: 39.22m | eta: 71.8m step 01975/05568 (35.47%) | loss: 2.656567 | lrm: 0.99 | dt: 1168.10ms | tok/sec: 897,677 | bf16_mfu: 55.14 | total time: 39.24m | eta: 71.7m step 01976/05568 (35.49%) | loss: 2.656520 | lrm: 0.99 | dt: 1181.37ms | tok/sec: 887,591 | bf16_mfu: 54.52 | total time: 39.26m | eta: 71.7m step 01977/05568 (35.51%) | loss: 2.662686 | lrm: 0.99 | dt: 1187.64ms | tok/sec: 882,907 | bf16_mfu: 54.23 | total time: 39.28m | eta: 71.7m step 01978/05568 (35.52%) | loss: 2.663403 | lrm: 0.99 | dt: 1185.79ms | tok/sec: 884,287 | bf16_mfu: 54.32 | total time: 39.30m | eta: 71.7m step 01979/05568 (35.54%) | loss: 2.663881 | lrm: 0.99 | dt: 1199.55ms | tok/sec: 874,143 | bf16_mfu: 53.70 | total time: 39.32m | eta: 71.7m step 01980/05568 (35.56%) | loss: 2.663558 | lrm: 0.99 | dt: 1196.90ms | tok/sec: 876,075 | bf16_mfu: 53.82 | total time: 39.34m | eta: 71.6m step 01981/05568 (35.58%) | loss: 2.673553 | lrm: 0.99 | dt: 1190.63ms | tok/sec: 880,688 | bf16_mfu: 54.10 | total time: 39.36m | eta: 71.6m step 01982/05568 (35.60%) | loss: 2.673823 | lrm: 0.99 | dt: 1195.14ms | tok/sec: 877,363 | bf16_mfu: 53.89 | total time: 39.38m | eta: 71.6m step 01983/05568 (35.61%) | loss: 2.677906 | lrm: 0.99 | dt: 1192.01ms | tok/sec: 879,668 | bf16_mfu: 54.04 | total time: 39.40m | eta: 71.6m step 01984/05568 (35.63%) | loss: 2.669200 | lrm: 0.99 | dt: 1198.33ms | tok/sec: 875,034 | bf16_mfu: 53.75 | total time: 39.42m | eta: 71.6m step 01985/05568 (35.65%) | loss: 2.662258 | lrm: 0.99 | dt: 1191.51ms | tok/sec: 880,038 | bf16_mfu: 54.06 | total time: 39.44m | eta: 71.5m step 01986/05568 (35.67%) | loss: 2.659112 | lrm: 0.99 | dt: 1187.52ms | tok/sec: 882,993 | bf16_mfu: 54.24 | total time: 39.46m | eta: 71.5m step 01987/05568 (35.69%) | loss: 2.672193 | lrm: 0.99 | dt: 1192.82ms | tok/sec: 879,072 | bf16_mfu: 54.00 | total time: 39.48m | eta: 71.5m step 01988/05568 (35.70%) | loss: 2.671902 | lrm: 0.99 | dt: 1198.26ms | tok/sec: 875,079 | bf16_mfu: 53.75 | total time: 39.50m | eta: 71.5m step 01989/05568 (35.72%) | loss: 2.668306 | lrm: 0.99 | dt: 1198.81ms | tok/sec: 874,680 | bf16_mfu: 53.73 | total time: 39.52m | eta: 71.5m step 01990/05568 (35.74%) | loss: 2.673209 | lrm: 0.99 | dt: 1194.72ms | tok/sec: 877,675 | bf16_mfu: 53.91 | total time: 39.54m | eta: 71.4m step 01991/05568 (35.76%) | loss: 2.672462 | lrm: 0.99 | dt: 1194.28ms | tok/sec: 878,000 | bf16_mfu: 53.93 | total time: 39.56m | eta: 71.4m step 01992/05568 (35.78%) | loss: 2.664735 | lrm: 0.99 | dt: 1202.91ms | tok/sec: 871,702 | bf16_mfu: 53.55 | total time: 39.58m | eta: 71.4m step 01993/05568 (35.79%) | loss: 2.675201 | lrm: 0.99 | dt: 1192.29ms | tok/sec: 879,464 | bf16_mfu: 54.02 | total time: 39.60m | eta: 71.4m step 01994/05568 (35.81%) | loss: 2.674674 | lrm: 0.99 | dt: 1191.02ms | tok/sec: 880,402 | bf16_mfu: 54.08 | total time: 39.62m | eta: 71.4m step 01995/05568 (35.83%) | loss: 2.668553 | lrm: 0.99 | dt: 1196.32ms | tok/sec: 876,497 | bf16_mfu: 53.84 | total time: 39.64m | eta: 71.3m step 01996/05568 (35.85%) | loss: 2.673149 | lrm: 0.99 | dt: 1201.77ms | tok/sec: 872,528 | bf16_mfu: 53.60 | total time: 39.66m | eta: 71.3m step 01997/05568 (35.87%) | loss: 2.680649 | lrm: 0.99 | dt: 1195.97ms | tok/sec: 876,754 | bf16_mfu: 53.86 | total time: 39.68m | eta: 71.3m step 01998/05568 (35.88%) | loss: 2.674607 | lrm: 0.99 | dt: 1192.44ms | tok/sec: 879,355 | bf16_mfu: 54.02 | total time: 39.70m | eta: 71.3m step 01999/05568 (35.90%) | loss: 2.665850 | lrm: 0.99 | dt: 1195.74ms | tok/sec: 876,924 | bf16_mfu: 53.87 | total time: 39.72m | eta: 71.3m step:2000/5568 val_bpb:0.809199 val_time:3.89s step 02000/05568 (35.92%) | loss: 2.672754 | lrm: 0.99 | dt: 1200.79ms | tok/sec: 873,239 | bf16_mfu: 53.64 | total time: 39.74m | eta: 71.2m step 02001/05568 (35.94%) | loss: 2.668365 | lrm: 0.99 | dt: 1193.72ms | tok/sec: 878,407 | bf16_mfu: 53.96 | total time: 39.76m | eta: 71.2m step 02002/05568 (35.96%) | loss: 2.670145 | lrm: 0.99 | dt: 1196.49ms | tok/sec: 876,373 | bf16_mfu: 53.83 | total time: 39.78m | eta: 71.2m step 02003/05568 (35.97%) | loss: 2.672920 | lrm: 0.99 | dt: 1196.41ms | tok/sec: 876,436 | bf16_mfu: 53.84 | total time: 39.80m | eta: 71.2m step 02004/05568 (35.99%) | loss: 2.683794 | lrm: 0.99 | dt: 1195.80ms | tok/sec: 876,881 | bf16_mfu: 53.86 | total time: 39.82m | eta: 71.2m step 02005/05568 (36.01%) | loss: 2.683089 | lrm: 0.99 | dt: 1195.35ms | tok/sec: 877,210 | bf16_mfu: 53.89 | total time: 39.84m | eta: 71.1m step 02006/05568 (36.03%) | loss: 2.684973 | lrm: 0.99 | dt: 1194.12ms | tok/sec: 878,116 | bf16_mfu: 53.94 | total time: 39.86m | eta: 71.1m step 02007/05568 (36.05%) | loss: 2.678488 | lrm: 0.98 | dt: 1196.73ms | tok/sec: 876,202 | bf16_mfu: 53.82 | total time: 39.88m | eta: 71.1m step 02008/05568 (36.06%) | loss: 2.671089 | lrm: 0.98 | dt: 1192.77ms | tok/sec: 879,107 | bf16_mfu: 54.00 | total time: 39.90m | eta: 71.1m step 02009/05568 (36.08%) | loss: 2.663475 | lrm: 0.98 | dt: 1192.93ms | tok/sec: 878,990 | bf16_mfu: 53.99 | total time: 39.92m | eta: 71.1m step 02010/05568 (36.10%) | loss: 2.660457 | lrm: 0.98 | dt: 1196.74ms | tok/sec: 876,190 | bf16_mfu: 53.82 | total time: 39.94m | eta: 71.0m step 02011/05568 (36.12%) | loss: 2.657814 | lrm: 0.98 | dt: 1198.26ms | tok/sec: 875,085 | bf16_mfu: 53.75 | total time: 39.96m | eta: 71.0m step 02012/05568 (36.14%) | loss: 2.658146 | lrm: 0.98 | dt: 1195.34ms | tok/sec: 877,222 | bf16_mfu: 53.89 | total time: 39.98m | eta: 71.0m step 02013/05568 (36.15%) | loss: 2.655182 | lrm: 0.98 | dt: 1193.69ms | tok/sec: 878,430 | bf16_mfu: 53.96 | total time: 40.00m | eta: 71.0m step 02014/05568 (36.17%) | loss: 2.656074 | lrm: 0.98 | dt: 1197.68ms | tok/sec: 875,506 | bf16_mfu: 53.78 | total time: 40.02m | eta: 71.0m step 02015/05568 (36.19%) | loss: 2.660353 | lrm: 0.98 | dt: 1198.69ms | tok/sec: 874,765 | bf16_mfu: 53.73 | total time: 40.04m | eta: 70.9m step 02016/05568 (36.21%) | loss: 2.657777 | lrm: 0.98 | dt: 1199.14ms | tok/sec: 874,437 | bf16_mfu: 53.71 | total time: 40.06m | eta: 70.9m step 02017/05568 (36.22%) | loss: 2.660112 | lrm: 0.98 | dt: 1197.06ms | tok/sec: 875,955 | bf16_mfu: 53.81 | total time: 40.08m | eta: 70.9m step 02018/05568 (36.24%) | loss: 2.660600 | lrm: 0.98 | dt: 1193.50ms | tok/sec: 878,575 | bf16_mfu: 53.97 | total time: 40.10m | eta: 70.9m step 02019/05568 (36.26%) | loss: 2.651062 | lrm: 0.98 | dt: 1200.10ms | tok/sec: 873,740 | bf16_mfu: 53.67 | total time: 40.12m | eta: 70.9m step 02020/05568 (36.28%) | loss: 2.652604 | lrm: 0.98 | dt: 1198.91ms | tok/sec: 874,605 | bf16_mfu: 53.73 | total time: 40.14m | eta: 70.8m step 02021/05568 (36.30%) | loss: 2.659391 | lrm: 0.98 | dt: 1196.94ms | tok/sec: 876,048 | bf16_mfu: 53.81 | total time: 40.16m | eta: 70.8m step 02022/05568 (36.31%) | loss: 2.672508 | lrm: 0.98 | dt: 1200.80ms | tok/sec: 873,231 | bf16_mfu: 53.64 | total time: 40.18m | eta: 70.8m step 02023/05568 (36.33%) | loss: 2.665583 | lrm: 0.98 | dt: 1194.25ms | tok/sec: 878,022 | bf16_mfu: 53.93 | total time: 40.20m | eta: 70.8m step 02024/05568 (36.35%) | loss: 2.665920 | lrm: 0.98 | dt: 1196.45ms | tok/sec: 876,404 | bf16_mfu: 53.84 | total time: 40.22m | eta: 70.8m step 02025/05568 (36.37%) | loss: 2.661286 | lrm: 0.98 | dt: 1198.45ms | tok/sec: 874,940 | bf16_mfu: 53.75 | total time: 40.24m | eta: 70.7m step 02026/05568 (36.39%) | loss: 2.652590 | lrm: 0.98 | dt: 1199.80ms | tok/sec: 873,961 | bf16_mfu: 53.69 | total time: 40.26m | eta: 70.7m step 02027/05568 (36.40%) | loss: 2.666412 | lrm: 0.98 | dt: 1192.13ms | tok/sec: 879,582 | bf16_mfu: 54.03 | total time: 40.28m | eta: 70.7m step 02028/05568 (36.42%) | loss: 2.661389 | lrm: 0.98 | dt: 1197.45ms | tok/sec: 875,676 | bf16_mfu: 53.79 | total time: 40.30m | eta: 70.7m step 02029/05568 (36.44%) | loss: 2.665052 | lrm: 0.98 | dt: 1203.75ms | tok/sec: 871,093 | bf16_mfu: 53.51 | total time: 40.32m | eta: 70.7m step 02030/05568 (36.46%) | loss: 2.671228 | lrm: 0.98 | dt: 1196.26ms | tok/sec: 876,545 | bf16_mfu: 53.84 | total time: 40.34m | eta: 70.6m step 02031/05568 (36.48%) | loss: 2.675539 | lrm: 0.98 | dt: 1196.68ms | tok/sec: 876,240 | bf16_mfu: 53.83 | total time: 40.36m | eta: 70.6m step 02032/05568 (36.49%) | loss: 2.678643 | lrm: 0.98 | dt: 1196.51ms | tok/sec: 876,358 | bf16_mfu: 53.83 | total time: 40.38m | eta: 70.6m step 02033/05568 (36.51%) | loss: 2.684769 | lrm: 0.98 | dt: 1199.88ms | tok/sec: 873,900 | bf16_mfu: 53.68 | total time: 40.40m | eta: 70.6m step 02034/05568 (36.53%) | loss: 2.682762 | lrm: 0.98 | dt: 1203.87ms | tok/sec: 871,001 | bf16_mfu: 53.50 | total time: 40.42m | eta: 70.6m step 02035/05568 (36.55%) | loss: 2.683857 | lrm: 0.98 | dt: 1198.74ms | tok/sec: 874,728 | bf16_mfu: 53.73 | total time: 40.44m | eta: 70.5m step 02036/05568 (36.57%) | loss: 2.691582 | lrm: 0.98 | dt: 1193.72ms | tok/sec: 878,407 | bf16_mfu: 53.96 | total time: 40.46m | eta: 70.5m step 02037/05568 (36.58%) | loss: 2.694362 | lrm: 0.98 | dt: 1189.28ms | tok/sec: 881,691 | bf16_mfu: 54.16 | total time: 40.48m | eta: 70.5m step 02038/05568 (36.60%) | loss: 2.684945 | lrm: 0.98 | dt: 1193.89ms | tok/sec: 878,288 | bf16_mfu: 53.95 | total time: 40.50m | eta: 70.5m step 02039/05568 (36.62%) | loss: 2.688860 | lrm: 0.98 | dt: 1203.69ms | tok/sec: 871,136 | bf16_mfu: 53.51 | total time: 40.52m | eta: 70.5m step 02040/05568 (36.64%) | loss: 2.684480 | lrm: 0.98 | dt: 1197.21ms | tok/sec: 875,846 | bf16_mfu: 53.80 | total time: 40.54m | eta: 70.4m step 02041/05568 (36.66%) | loss: 2.676469 | lrm: 0.98 | dt: 1198.08ms | tok/sec: 875,215 | bf16_mfu: 53.76 | total time: 40.56m | eta: 70.4m step 02042/05568 (36.67%) | loss: 2.682409 | lrm: 0.98 | dt: 1199.43ms | tok/sec: 874,228 | bf16_mfu: 53.70 | total time: 40.58m | eta: 70.4m step 02043/05568 (36.69%) | loss: 2.690764 | lrm: 0.98 | dt: 1199.91ms | tok/sec: 873,877 | bf16_mfu: 53.68 | total time: 40.60m | eta: 70.4m step 02044/05568 (36.71%) | loss: 2.684864 | lrm: 0.98 | dt: 1194.95ms | tok/sec: 877,503 | bf16_mfu: 53.90 | total time: 40.62m | eta: 70.4m step 02045/05568 (36.73%) | loss: 2.673158 | lrm: 0.97 | dt: 1202.43ms | tok/sec: 872,048 | bf16_mfu: 53.57 | total time: 40.64m | eta: 70.3m step 02046/05568 (36.75%) | loss: 2.685090 | lrm: 0.97 | dt: 1197.07ms | tok/sec: 875,949 | bf16_mfu: 53.81 | total time: 40.66m | eta: 70.3m step 02047/05568 (36.76%) | loss: 2.693247 | lrm: 0.97 | dt: 1201.78ms | tok/sec: 872,518 | bf16_mfu: 53.60 | total time: 40.68m | eta: 70.3m step 02048/05568 (36.78%) | loss: 2.680083 | lrm: 0.97 | dt: 1197.92ms | tok/sec: 875,328 | bf16_mfu: 53.77 | total time: 40.70m | eta: 70.3m step 02049/05568 (36.80%) | loss: 2.668422 | lrm: 0.97 | dt: 1196.48ms | tok/sec: 876,385 | bf16_mfu: 53.83 | total time: 40.72m | eta: 70.3m step 02050/05568 (36.82%) | loss: 2.664279 | lrm: 0.97 | dt: 1197.33ms | tok/sec: 875,764 | bf16_mfu: 53.80 | total time: 40.74m | eta: 70.2m step 02051/05568 (36.84%) | loss: 2.654203 | lrm: 0.97 | dt: 1194.15ms | tok/sec: 878,094 | bf16_mfu: 53.94 | total time: 40.76m | eta: 70.2m step 02052/05568 (36.85%) | loss: 2.654655 | lrm: 0.97 | dt: 1198.65ms | tok/sec: 874,796 | bf16_mfu: 53.74 | total time: 40.78m | eta: 70.2m step 02053/05568 (36.87%) | loss: 2.652365 | lrm: 0.97 | dt: 1197.94ms | tok/sec: 875,317 | bf16_mfu: 53.77 | total time: 40.80m | eta: 70.2m step 02054/05568 (36.89%) | loss: 2.653269 | lrm: 0.97 | dt: 1194.08ms | tok/sec: 878,144 | bf16_mfu: 53.94 | total time: 40.82m | eta: 70.2m step 02055/05568 (36.91%) | loss: 2.655387 | lrm: 0.97 | dt: 1196.37ms | tok/sec: 876,467 | bf16_mfu: 53.84 | total time: 40.84m | eta: 70.1m step 02056/05568 (36.93%) | loss: 2.660176 | lrm: 0.97 | dt: 1201.38ms | tok/sec: 872,811 | bf16_mfu: 53.61 | total time: 40.86m | eta: 70.1m step 02057/05568 (36.94%) | loss: 2.665616 | lrm: 0.97 | dt: 1195.17ms | tok/sec: 877,345 | bf16_mfu: 53.89 | total time: 40.88m | eta: 70.1m step 02058/05568 (36.96%) | loss: 2.667859 | lrm: 0.97 | dt: 1200.75ms | tok/sec: 873,271 | bf16_mfu: 53.64 | total time: 40.90m | eta: 70.1m step 02059/05568 (36.98%) | loss: 2.669958 | lrm: 0.97 | dt: 1197.80ms | tok/sec: 875,416 | bf16_mfu: 53.77 | total time: 40.92m | eta: 70.1m step 02060/05568 (37.00%) | loss: 2.663889 | lrm: 0.97 | dt: 1205.93ms | tok/sec: 869,513 | bf16_mfu: 53.41 | total time: 40.94m | eta: 70.0m step 02061/05568 (37.02%) | loss: 2.660416 | lrm: 0.97 | dt: 1199.61ms | tok/sec: 874,098 | bf16_mfu: 53.69 | total time: 40.96m | eta: 70.0m step 02062/05568 (37.03%) | loss: 2.656391 | lrm: 0.97 | dt: 1194.50ms | tok/sec: 877,835 | bf16_mfu: 53.92 | total time: 40.98m | eta: 70.0m step 02063/05568 (37.05%) | loss: 2.658345 | lrm: 0.97 | dt: 1203.49ms | tok/sec: 871,280 | bf16_mfu: 53.52 | total time: 41.00m | eta: 70.0m step 02064/05568 (37.07%) | loss: 2.661209 | lrm: 0.97 | dt: 1200.34ms | tok/sec: 873,568 | bf16_mfu: 53.66 | total time: 41.02m | eta: 70.0m step 02065/05568 (37.09%) | loss: 2.665271 | lrm: 0.97 | dt: 1195.19ms | tok/sec: 877,327 | bf16_mfu: 53.89 | total time: 41.04m | eta: 69.9m step 02066/05568 (37.10%) | loss: 2.665898 | lrm: 0.97 | dt: 1198.85ms | tok/sec: 874,652 | bf16_mfu: 53.73 | total time: 41.06m | eta: 69.9m step 02067/05568 (37.12%) | loss: 2.658408 | lrm: 0.97 | dt: 1198.04ms | tok/sec: 875,243 | bf16_mfu: 53.76 | total time: 41.08m | eta: 69.9m step 02068/05568 (37.14%) | loss: 2.651184 | lrm: 0.97 | dt: 1200.31ms | tok/sec: 873,588 | bf16_mfu: 53.66 | total time: 41.10m | eta: 69.9m step 02069/05568 (37.16%) | loss: 2.641166 | lrm: 0.97 | dt: 1202.54ms | tok/sec: 871,968 | bf16_mfu: 53.56 | total time: 41.12m | eta: 69.9m step 02070/05568 (37.18%) | loss: 2.642409 | lrm: 0.97 | dt: 1196.37ms | tok/sec: 876,463 | bf16_mfu: 53.84 | total time: 41.14m | eta: 69.8m step 02071/05568 (37.19%) | loss: 2.649073 | lrm: 0.97 | dt: 1202.86ms | tok/sec: 871,735 | bf16_mfu: 53.55 | total time: 41.16m | eta: 69.8m step 02072/05568 (37.21%) | loss: 2.648670 | lrm: 0.97 | dt: 1198.45ms | tok/sec: 874,943 | bf16_mfu: 53.75 | total time: 41.18m | eta: 69.8m step 02073/05568 (37.23%) | loss: 2.653257 | lrm: 0.97 | dt: 1201.42ms | tok/sec: 872,778 | bf16_mfu: 53.61 | total time: 41.20m | eta: 69.8m step 02074/05568 (37.25%) | loss: 2.661336 | lrm: 0.97 | dt: 1199.24ms | tok/sec: 874,364 | bf16_mfu: 53.71 | total time: 41.22m | eta: 69.8m step 02075/05568 (37.27%) | loss: 2.667602 | lrm: 0.97 | dt: 1199.91ms | tok/sec: 873,876 | bf16_mfu: 53.68 | total time: 41.24m | eta: 69.7m step 02076/05568 (37.28%) | loss: 2.663973 | lrm: 0.97 | dt: 1200.01ms | tok/sec: 873,804 | bf16_mfu: 53.68 | total time: 41.26m | eta: 69.7m step 02077/05568 (37.30%) | loss: 2.671512 | lrm: 0.97 | dt: 1201.94ms | tok/sec: 872,403 | bf16_mfu: 53.59 | total time: 41.28m | eta: 69.7m step 02078/05568 (37.32%) | loss: 2.669366 | lrm: 0.97 | dt: 1198.44ms | tok/sec: 874,947 | bf16_mfu: 53.75 | total time: 41.30m | eta: 69.7m step 02079/05568 (37.34%) | loss: 2.671085 | lrm: 0.97 | dt: 1200.43ms | tok/sec: 873,497 | bf16_mfu: 53.66 | total time: 41.32m | eta: 69.7m step 02080/05568 (37.36%) | loss: 2.681230 | lrm: 0.97 | dt: 1196.49ms | tok/sec: 876,379 | bf16_mfu: 53.83 | total time: 41.34m | eta: 69.6m step 02081/05568 (37.37%) | loss: 2.680126 | lrm: 0.97 | dt: 1203.59ms | tok/sec: 871,208 | bf16_mfu: 53.52 | total time: 41.36m | eta: 69.6m step 02082/05568 (37.39%) | loss: 2.680668 | lrm: 0.97 | dt: 1199.09ms | tok/sec: 874,474 | bf16_mfu: 53.72 | total time: 41.38m | eta: 69.6m step 02083/05568 (37.41%) | loss: 2.668412 | lrm: 0.96 | dt: 1196.17ms | tok/sec: 876,608 | bf16_mfu: 53.85 | total time: 41.40m | eta: 69.6m step 02084/05568 (37.43%) | loss: 2.661073 | lrm: 0.96 | dt: 1195.61ms | tok/sec: 877,023 | bf16_mfu: 53.87 | total time: 41.42m | eta: 69.6m step 02085/05568 (37.45%) | loss: 2.650658 | lrm: 0.96 | dt: 1196.67ms | tok/sec: 876,243 | bf16_mfu: 53.83 | total time: 41.44m | eta: 69.5m step 02086/05568 (37.46%) | loss: 2.646556 | lrm: 0.96 | dt: 1196.99ms | tok/sec: 876,011 | bf16_mfu: 53.81 | total time: 41.46m | eta: 69.5m step 02087/05568 (37.48%) | loss: 2.649536 | lrm: 0.96 | dt: 1195.43ms | tok/sec: 877,152 | bf16_mfu: 53.88 | total time: 41.48m | eta: 69.5m step 02088/05568 (37.50%) | loss: 2.636750 | lrm: 0.96 | dt: 1205.09ms | tok/sec: 870,120 | bf16_mfu: 53.45 | total time: 41.50m | eta: 69.5m step 02089/05568 (37.52%) | loss: 2.634600 | lrm: 0.96 | dt: 1194.95ms | tok/sec: 877,506 | bf16_mfu: 53.90 | total time: 41.52m | eta: 69.5m step 02090/05568 (37.54%) | loss: 2.648619 | lrm: 0.96 | dt: 1202.83ms | tok/sec: 871,757 | bf16_mfu: 53.55 | total time: 41.54m | eta: 69.4m step 02091/05568 (37.55%) | loss: 2.641755 | lrm: 0.96 | dt: 1199.22ms | tok/sec: 874,381 | bf16_mfu: 53.71 | total time: 41.56m | eta: 69.4m step 02092/05568 (37.57%) | loss: 2.648503 | lrm: 0.96 | dt: 1199.44ms | tok/sec: 874,222 | bf16_mfu: 53.70 | total time: 41.58m | eta: 69.4m step 02093/05568 (37.59%) | loss: 2.659264 | lrm: 0.96 | dt: 1198.31ms | tok/sec: 875,046 | bf16_mfu: 53.75 | total time: 41.60m | eta: 69.4m step 02094/05568 (37.61%) | loss: 2.659031 | lrm: 0.96 | dt: 1194.83ms | tok/sec: 877,591 | bf16_mfu: 53.91 | total time: 41.62m | eta: 69.4m step 02095/05568 (37.63%) | loss: 2.650716 | lrm: 0.96 | dt: 1207.85ms | tok/sec: 868,131 | bf16_mfu: 53.33 | total time: 41.64m | eta: 69.3m step 02096/05568 (37.64%) | loss: 2.645193 | lrm: 0.96 | dt: 1198.38ms | tok/sec: 874,997 | bf16_mfu: 53.75 | total time: 41.66m | eta: 69.3m step 02097/05568 (37.66%) | loss: 2.635465 | lrm: 0.96 | dt: 1197.61ms | tok/sec: 875,553 | bf16_mfu: 53.78 | total time: 41.68m | eta: 69.3m step 02098/05568 (37.68%) | loss: 2.629318 | lrm: 0.96 | dt: 1198.70ms | tok/sec: 874,759 | bf16_mfu: 53.73 | total time: 41.70m | eta: 69.3m step 02099/05568 (37.70%) | loss: 2.626687 | lrm: 0.96 | dt: 1205.77ms | tok/sec: 869,629 | bf16_mfu: 53.42 | total time: 41.72m | eta: 69.3m step 02100/05568 (37.72%) | loss: 2.637999 | lrm: 0.96 | dt: 1196.93ms | tok/sec: 876,054 | bf16_mfu: 53.81 | total time: 41.74m | eta: 69.2m step 02101/05568 (37.73%) | loss: 2.648308 | lrm: 0.96 | dt: 1198.21ms | tok/sec: 875,119 | bf16_mfu: 53.76 | total time: 41.76m | eta: 69.2m step 02102/05568 (37.75%) | loss: 2.646731 | lrm: 0.96 | dt: 1204.14ms | tok/sec: 870,806 | bf16_mfu: 53.49 | total time: 41.78m | eta: 69.2m step 02103/05568 (37.77%) | loss: 2.649050 | lrm: 0.96 | dt: 1207.36ms | tok/sec: 868,487 | bf16_mfu: 53.35 | total time: 41.80m | eta: 69.2m step 02104/05568 (37.79%) | loss: 2.644609 | lrm: 0.96 | dt: 1197.06ms | tok/sec: 875,959 | bf16_mfu: 53.81 | total time: 41.82m | eta: 69.2m step 02105/05568 (37.81%) | loss: 2.643413 | lrm: 0.96 | dt: 1198.64ms | tok/sec: 874,803 | bf16_mfu: 53.74 | total time: 41.84m | eta: 69.1m step 02106/05568 (37.82%) | loss: 2.640757 | lrm: 0.96 | dt: 1198.21ms | tok/sec: 875,116 | bf16_mfu: 53.76 | total time: 41.86m | eta: 69.1m step 02107/05568 (37.84%) | loss: 2.650104 | lrm: 0.96 | dt: 1200.97ms | tok/sec: 873,108 | bf16_mfu: 53.63 | total time: 41.88m | eta: 69.1m step 02108/05568 (37.86%) | loss: 2.651433 | lrm: 0.96 | dt: 1195.98ms | tok/sec: 876,746 | bf16_mfu: 53.86 | total time: 41.90m | eta: 69.1m step 02109/05568 (37.88%) | loss: 2.645680 | lrm: 0.96 | dt: 1201.49ms | tok/sec: 872,726 | bf16_mfu: 53.61 | total time: 41.92m | eta: 69.1m step 02110/05568 (37.90%) | loss: 2.649036 | lrm: 0.96 | dt: 1199.01ms | tok/sec: 874,532 | bf16_mfu: 53.72 | total time: 41.94m | eta: 69.0m step 02111/05568 (37.91%) | loss: 2.654477 | lrm: 0.96 | dt: 1196.06ms | tok/sec: 876,695 | bf16_mfu: 53.85 | total time: 41.96m | eta: 69.0m step 02112/05568 (37.93%) | loss: 2.660775 | lrm: 0.96 | dt: 1199.65ms | tok/sec: 874,069 | bf16_mfu: 53.69 | total time: 41.98m | eta: 69.0m step 02113/05568 (37.95%) | loss: 2.655141 | lrm: 0.96 | dt: 1206.85ms | tok/sec: 868,855 | bf16_mfu: 53.37 | total time: 42.00m | eta: 69.0m step 02114/05568 (37.97%) | loss: 2.661562 | lrm: 0.96 | dt: 1195.30ms | tok/sec: 877,251 | bf16_mfu: 53.89 | total time: 42.02m | eta: 69.0m step 02115/05568 (37.98%) | loss: 2.654284 | lrm: 0.96 | dt: 1199.15ms | tok/sec: 874,432 | bf16_mfu: 53.71 | total time: 42.04m | eta: 68.9m step 02116/05568 (38.00%) | loss: 2.662929 | lrm: 0.96 | dt: 1207.79ms | tok/sec: 868,175 | bf16_mfu: 53.33 | total time: 42.06m | eta: 68.9m step 02117/05568 (38.02%) | loss: 2.653860 | lrm: 0.96 | dt: 1202.82ms | tok/sec: 871,766 | bf16_mfu: 53.55 | total time: 42.08m | eta: 68.9m step 02118/05568 (38.04%) | loss: 2.653718 | lrm: 0.96 | dt: 1196.39ms | tok/sec: 876,448 | bf16_mfu: 53.84 | total time: 42.10m | eta: 68.9m step 02119/05568 (38.06%) | loss: 2.655350 | lrm: 0.96 | dt: 1198.56ms | tok/sec: 874,862 | bf16_mfu: 53.74 | total time: 42.12m | eta: 68.9m step 02120/05568 (38.07%) | loss: 2.652227 | lrm: 0.96 | dt: 1195.18ms | tok/sec: 877,338 | bf16_mfu: 53.89 | total time: 42.14m | eta: 68.8m step 02121/05568 (38.09%) | loss: 2.654430 | lrm: 0.95 | dt: 1199.52ms | tok/sec: 874,164 | bf16_mfu: 53.70 | total time: 42.16m | eta: 68.8m step 02122/05568 (38.11%) | loss: 2.638652 | lrm: 0.95 | dt: 1198.28ms | tok/sec: 875,067 | bf16_mfu: 53.75 | total time: 42.18m | eta: 68.8m step 02123/05568 (38.13%) | loss: 2.639136 | lrm: 0.95 | dt: 1199.15ms | tok/sec: 874,433 | bf16_mfu: 53.71 | total time: 42.20m | eta: 68.8m step 02124/05568 (38.15%) | loss: 2.634113 | lrm: 0.95 | dt: 1200.47ms | tok/sec: 873,472 | bf16_mfu: 53.66 | total time: 42.22m | eta: 68.8m step 02125/05568 (38.16%) | loss: 2.640948 | lrm: 0.95 | dt: 1200.35ms | tok/sec: 873,557 | bf16_mfu: 53.66 | total time: 42.24m | eta: 68.7m step 02126/05568 (38.18%) | loss: 2.644360 | lrm: 0.95 | dt: 1199.05ms | tok/sec: 874,502 | bf16_mfu: 53.72 | total time: 42.26m | eta: 68.7m step 02127/05568 (38.20%) | loss: 2.640822 | lrm: 0.95 | dt: 1195.06ms | tok/sec: 877,426 | bf16_mfu: 53.90 | total time: 42.28m | eta: 68.7m step 02128/05568 (38.22%) | loss: 2.646245 | lrm: 0.95 | dt: 1198.15ms | tok/sec: 875,165 | bf16_mfu: 53.76 | total time: 42.30m | eta: 68.7m step 02129/05568 (38.24%) | loss: 2.648252 | lrm: 0.95 | dt: 1196.94ms | tok/sec: 876,048 | bf16_mfu: 53.81 | total time: 42.32m | eta: 68.7m step 02130/05568 (38.25%) | loss: 2.656712 | lrm: 0.95 | dt: 1200.44ms | tok/sec: 873,490 | bf16_mfu: 53.66 | total time: 42.34m | eta: 68.6m step 02131/05568 (38.27%) | loss: 2.663673 | lrm: 0.95 | dt: 1201.92ms | tok/sec: 872,416 | bf16_mfu: 53.59 | total time: 42.36m | eta: 68.6m step 02132/05568 (38.29%) | loss: 2.660583 | lrm: 0.95 | dt: 1196.68ms | tok/sec: 876,234 | bf16_mfu: 53.83 | total time: 42.38m | eta: 68.6m step 02133/05568 (38.31%) | loss: 2.655573 | lrm: 0.95 | dt: 1201.49ms | tok/sec: 872,727 | bf16_mfu: 53.61 | total time: 42.40m | eta: 68.6m step 02134/05568 (38.33%) | loss: 2.647300 | lrm: 0.95 | dt: 1198.86ms | tok/sec: 874,644 | bf16_mfu: 53.73 | total time: 42.42m | eta: 68.6m step 02135/05568 (38.34%) | loss: 2.643505 | lrm: 0.95 | dt: 1198.04ms | tok/sec: 875,241 | bf16_mfu: 53.76 | total time: 42.44m | eta: 68.5m step 02136/05568 (38.36%) | loss: 2.643524 | lrm: 0.95 | dt: 1196.60ms | tok/sec: 876,293 | bf16_mfu: 53.83 | total time: 42.46m | eta: 68.5m step 02137/05568 (38.38%) | loss: 2.635382 | lrm: 0.95 | dt: 1199.46ms | tok/sec: 874,207 | bf16_mfu: 53.70 | total time: 42.48m | eta: 68.5m step 02138/05568 (38.40%) | loss: 2.632114 | lrm: 0.95 | dt: 1194.89ms | tok/sec: 877,549 | bf16_mfu: 53.91 | total time: 42.50m | eta: 68.5m step 02139/05568 (38.42%) | loss: 2.643412 | lrm: 0.95 | dt: 1195.93ms | tok/sec: 876,784 | bf16_mfu: 53.86 | total time: 42.52m | eta: 68.5m step 02140/05568 (38.43%) | loss: 2.646462 | lrm: 0.95 | dt: 1198.15ms | tok/sec: 875,159 | bf16_mfu: 53.76 | total time: 42.54m | eta: 68.4m step 02141/05568 (38.45%) | loss: 2.650703 | lrm: 0.95 | dt: 1197.42ms | tok/sec: 875,699 | bf16_mfu: 53.79 | total time: 42.56m | eta: 68.4m step 02142/05568 (38.47%) | loss: 2.639030 | lrm: 0.95 | dt: 1199.95ms | tok/sec: 873,846 | bf16_mfu: 53.68 | total time: 42.58m | eta: 68.4m step 02143/05568 (38.49%) | loss: 2.630154 | lrm: 0.95 | dt: 1200.61ms | tok/sec: 873,368 | bf16_mfu: 53.65 | total time: 42.60m | eta: 68.4m step 02144/05568 (38.51%) | loss: 2.621599 | lrm: 0.95 | dt: 1198.60ms | tok/sec: 874,832 | bf16_mfu: 53.74 | total time: 42.62m | eta: 68.4m step 02145/05568 (38.52%) | loss: 2.624373 | lrm: 0.95 | dt: 1201.33ms | tok/sec: 872,844 | bf16_mfu: 53.62 | total time: 42.64m | eta: 68.3m step 02146/05568 (38.54%) | loss: 2.625498 | lrm: 0.95 | dt: 1195.97ms | tok/sec: 876,757 | bf16_mfu: 53.86 | total time: 42.66m | eta: 68.3m step 02147/05568 (38.56%) | loss: 2.630687 | lrm: 0.95 | dt: 1201.22ms | tok/sec: 872,922 | bf16_mfu: 53.62 | total time: 42.68m | eta: 68.3m step 02148/05568 (38.58%) | loss: 2.637009 | lrm: 0.95 | dt: 1201.12ms | tok/sec: 872,996 | bf16_mfu: 53.63 | total time: 42.70m | eta: 68.3m step 02149/05568 (38.60%) | loss: 2.644553 | lrm: 0.95 | dt: 1195.61ms | tok/sec: 877,025 | bf16_mfu: 53.87 | total time: 42.72m | eta: 68.3m step 02150/05568 (38.61%) | loss: 2.649836 | lrm: 0.95 | dt: 1198.51ms | tok/sec: 874,901 | bf16_mfu: 53.74 | total time: 42.74m | eta: 68.2m step 02151/05568 (38.63%) | loss: 2.645566 | lrm: 0.95 | dt: 1196.12ms | tok/sec: 876,647 | bf16_mfu: 53.85 | total time: 42.76m | eta: 68.2m step 02152/05568 (38.65%) | loss: 2.636102 | lrm: 0.95 | dt: 1189.00ms | tok/sec: 881,895 | bf16_mfu: 54.17 | total time: 42.78m | eta: 68.2m step 02153/05568 (38.67%) | loss: 2.641449 | lrm: 0.95 | dt: 1199.99ms | tok/sec: 873,822 | bf16_mfu: 53.68 | total time: 42.80m | eta: 68.2m step 02154/05568 (38.69%) | loss: 2.640297 | lrm: 0.95 | dt: 1194.27ms | tok/sec: 878,006 | bf16_mfu: 53.93 | total time: 42.82m | eta: 68.2m step 02155/05568 (38.70%) | loss: 2.645424 | lrm: 0.95 | dt: 1195.01ms | tok/sec: 877,460 | bf16_mfu: 53.90 | total time: 42.84m | eta: 68.1m step 02156/05568 (38.72%) | loss: 2.641379 | lrm: 0.95 | dt: 1188.96ms | tok/sec: 881,928 | bf16_mfu: 54.17 | total time: 42.86m | eta: 68.1m step 02157/05568 (38.74%) | loss: 2.647691 | lrm: 0.95 | dt: 1195.25ms | tok/sec: 877,284 | bf16_mfu: 53.89 | total time: 42.88m | eta: 68.1m step 02158/05568 (38.76%) | loss: 2.645943 | lrm: 0.95 | dt: 1197.85ms | tok/sec: 875,381 | bf16_mfu: 53.77 | total time: 42.90m | eta: 68.1m step 02159/05568 (38.78%) | loss: 2.650257 | lrm: 0.94 | dt: 1201.13ms | tok/sec: 872,994 | bf16_mfu: 53.63 | total time: 42.92m | eta: 68.1m step 02160/05568 (38.79%) | loss: 2.652006 | lrm: 0.94 | dt: 1198.52ms | tok/sec: 874,892 | bf16_mfu: 53.74 | total time: 42.94m | eta: 68.0m step 02161/05568 (38.81%) | loss: 2.654942 | lrm: 0.94 | dt: 1197.10ms | tok/sec: 875,931 | bf16_mfu: 53.81 | total time: 42.96m | eta: 68.0m step 02162/05568 (38.83%) | loss: 2.645584 | lrm: 0.94 | dt: 1196.77ms | tok/sec: 876,174 | bf16_mfu: 53.82 | total time: 42.98m | eta: 68.0m step 02163/05568 (38.85%) | loss: 2.648429 | lrm: 0.94 | dt: 1196.35ms | tok/sec: 876,477 | bf16_mfu: 53.84 | total time: 42.99m | eta: 68.0m step 02164/05568 (38.86%) | loss: 2.660068 | lrm: 0.94 | dt: 1203.22ms | tok/sec: 871,474 | bf16_mfu: 53.53 | total time: 43.02m | eta: 68.0m step 02165/05568 (38.88%) | loss: 2.673078 | lrm: 0.94 | dt: 1197.71ms | tok/sec: 875,481 | bf16_mfu: 53.78 | total time: 43.03m | eta: 67.9m step 02166/05568 (38.90%) | loss: 2.665632 | lrm: 0.94 | dt: 1203.52ms | tok/sec: 871,255 | bf16_mfu: 53.52 | total time: 43.06m | eta: 67.9m step 02167/05568 (38.92%) | loss: 2.654628 | lrm: 0.94 | dt: 1193.63ms | tok/sec: 878,477 | bf16_mfu: 53.96 | total time: 43.07m | eta: 67.9m step 02168/05568 (38.94%) | loss: 2.655885 | lrm: 0.94 | dt: 1197.35ms | tok/sec: 875,750 | bf16_mfu: 53.80 | total time: 43.09m | eta: 67.9m step 02169/05568 (38.95%) | loss: 2.649478 | lrm: 0.94 | dt: 1194.11ms | tok/sec: 878,119 | bf16_mfu: 53.94 | total time: 43.11m | eta: 67.9m step 02170/05568 (38.97%) | loss: 2.658038 | lrm: 0.94 | dt: 1199.69ms | tok/sec: 874,035 | bf16_mfu: 53.69 | total time: 43.13m | eta: 67.8m step 02171/05568 (38.99%) | loss: 2.658468 | lrm: 0.94 | dt: 1196.37ms | tok/sec: 876,465 | bf16_mfu: 53.84 | total time: 43.15m | eta: 67.8m step 02172/05568 (39.01%) | loss: 2.648412 | lrm: 0.94 | dt: 1199.11ms | tok/sec: 874,463 | bf16_mfu: 53.72 | total time: 43.17m | eta: 67.8m step 02173/05568 (39.03%) | loss: 2.642441 | lrm: 0.94 | dt: 1198.05ms | tok/sec: 875,237 | bf16_mfu: 53.76 | total time: 43.19m | eta: 67.8m step 02174/05568 (39.04%) | loss: 2.635650 | lrm: 0.94 | dt: 1193.33ms | tok/sec: 878,699 | bf16_mfu: 53.98 | total time: 43.21m | eta: 67.8m step 02175/05568 (39.06%) | loss: 2.638269 | lrm: 0.94 | dt: 1202.08ms | tok/sec: 872,299 | bf16_mfu: 53.58 | total time: 43.23m | eta: 67.7m step 02176/05568 (39.08%) | loss: 2.629514 | lrm: 0.94 | dt: 1194.40ms | tok/sec: 877,910 | bf16_mfu: 53.93 | total time: 43.25m | eta: 67.7m step 02177/05568 (39.10%) | loss: 2.630218 | lrm: 0.94 | dt: 1197.11ms | tok/sec: 875,920 | bf16_mfu: 53.81 | total time: 43.27m | eta: 67.7m step 02178/05568 (39.12%) | loss: 2.625365 | lrm: 0.94 | dt: 1201.92ms | tok/sec: 872,414 | bf16_mfu: 53.59 | total time: 43.29m | eta: 67.7m step 02179/05568 (39.13%) | loss: 2.632402 | lrm: 0.94 | dt: 1199.48ms | tok/sec: 874,193 | bf16_mfu: 53.70 | total time: 43.31m | eta: 67.7m step 02180/05568 (39.15%) | loss: 2.640987 | lrm: 0.94 | dt: 1198.26ms | tok/sec: 875,080 | bf16_mfu: 53.75 | total time: 43.33m | eta: 67.6m step 02181/05568 (39.17%) | loss: 2.638831 | lrm: 0.94 | dt: 1200.62ms | tok/sec: 873,359 | bf16_mfu: 53.65 | total time: 43.35m | eta: 67.6m step 02182/05568 (39.19%) | loss: 2.638693 | lrm: 0.94 | dt: 1199.15ms | tok/sec: 874,430 | bf16_mfu: 53.71 | total time: 43.37m | eta: 67.6m step 02183/05568 (39.21%) | loss: 2.635010 | lrm: 0.94 | dt: 1199.32ms | tok/sec: 874,310 | bf16_mfu: 53.71 | total time: 43.39m | eta: 67.6m step 02184/05568 (39.22%) | loss: 2.630639 | lrm: 0.94 | dt: 1192.37ms | tok/sec: 879,401 | bf16_mfu: 54.02 | total time: 43.41m | eta: 67.6m step 02185/05568 (39.24%) | loss: 2.637332 | lrm: 0.94 | dt: 1195.22ms | tok/sec: 877,309 | bf16_mfu: 53.89 | total time: 43.43m | eta: 67.5m step 02186/05568 (39.26%) | loss: 2.639969 | lrm: 0.94 | dt: 1193.41ms | tok/sec: 878,634 | bf16_mfu: 53.97 | total time: 43.45m | eta: 67.5m step 02187/05568 (39.28%) | loss: 2.646927 | lrm: 0.94 | dt: 1200.41ms | tok/sec: 873,515 | bf16_mfu: 53.66 | total time: 43.47m | eta: 67.5m step 02188/05568 (39.30%) | loss: 2.646425 | lrm: 0.94 | dt: 1198.69ms | tok/sec: 874,765 | bf16_mfu: 53.73 | total time: 43.49m | eta: 67.5m step 02189/05568 (39.31%) | loss: 2.644297 | lrm: 0.94 | dt: 1195.44ms | tok/sec: 877,148 | bf16_mfu: 53.88 | total time: 43.51m | eta: 67.5m step 02190/05568 (39.33%) | loss: 2.632076 | lrm: 0.94 | dt: 1194.46ms | tok/sec: 877,867 | bf16_mfu: 53.93 | total time: 43.53m | eta: 67.4m step 02191/05568 (39.35%) | loss: 2.635276 | lrm: 0.94 | dt: 1196.64ms | tok/sec: 876,264 | bf16_mfu: 53.83 | total time: 43.55m | eta: 67.4m step 02192/05568 (39.37%) | loss: 2.621614 | lrm: 0.94 | dt: 1192.80ms | tok/sec: 879,091 | bf16_mfu: 54.00 | total time: 43.57m | eta: 67.4m step 02193/05568 (39.39%) | loss: 2.636142 | lrm: 0.94 | dt: 1194.11ms | tok/sec: 878,124 | bf16_mfu: 53.94 | total time: 43.59m | eta: 67.4m step 02194/05568 (39.40%) | loss: 2.625838 | lrm: 0.94 | dt: 1195.49ms | tok/sec: 877,112 | bf16_mfu: 53.88 | total time: 43.61m | eta: 67.4m step 02195/05568 (39.42%) | loss: 2.634352 | lrm: 0.94 | dt: 1193.63ms | tok/sec: 878,473 | bf16_mfu: 53.96 | total time: 43.63m | eta: 67.3m step 02196/05568 (39.44%) | loss: 2.637478 | lrm: 0.94 | dt: 1199.77ms | tok/sec: 873,978 | bf16_mfu: 53.69 | total time: 43.65m | eta: 67.3m step 02197/05568 (39.46%) | loss: 2.644365 | lrm: 0.93 | dt: 1198.68ms | tok/sec: 874,774 | bf16_mfu: 53.74 | total time: 43.67m | eta: 67.3m step 02198/05568 (39.48%) | loss: 2.643175 | lrm: 0.93 | dt: 1201.87ms | tok/sec: 872,453 | bf16_mfu: 53.59 | total time: 43.69m | eta: 67.3m step 02199/05568 (39.49%) | loss: 2.635492 | lrm: 0.93 | dt: 1194.52ms | tok/sec: 877,824 | bf16_mfu: 53.92 | total time: 43.71m | eta: 67.3m step 02200/05568 (39.51%) | loss: 2.637710 | lrm: 0.93 | dt: 1197.06ms | tok/sec: 875,961 | bf16_mfu: 53.81 | total time: 43.73m | eta: 67.2m step 02201/05568 (39.53%) | loss: 2.629700 | lrm: 0.93 | dt: 1198.80ms | tok/sec: 874,686 | bf16_mfu: 53.73 | total time: 43.75m | eta: 67.2m step 02202/05568 (39.55%) | loss: 2.635495 | lrm: 0.93 | dt: 1199.11ms | tok/sec: 874,460 | bf16_mfu: 53.72 | total time: 43.77m | eta: 67.2m step 02203/05568 (39.57%) | loss: 2.639956 | lrm: 0.93 | dt: 1199.97ms | tok/sec: 873,833 | bf16_mfu: 53.68 | total time: 43.79m | eta: 67.2m step 02204/05568 (39.58%) | loss: 2.634081 | lrm: 0.93 | dt: 1195.28ms | tok/sec: 877,266 | bf16_mfu: 53.89 | total time: 43.81m | eta: 67.2m step 02205/05568 (39.60%) | loss: 2.623242 | lrm: 0.93 | dt: 1200.50ms | tok/sec: 873,447 | bf16_mfu: 53.65 | total time: 43.83m | eta: 67.1m step 02206/05568 (39.62%) | loss: 2.615575 | lrm: 0.93 | dt: 1199.95ms | tok/sec: 873,850 | bf16_mfu: 53.68 | total time: 43.85m | eta: 67.1m step 02207/05568 (39.64%) | loss: 2.604196 | lrm: 0.93 | dt: 1201.68ms | tok/sec: 872,592 | bf16_mfu: 53.60 | total time: 43.87m | eta: 67.1m step 02208/05568 (39.66%) | loss: 2.608362 | lrm: 0.93 | dt: 1197.23ms | tok/sec: 875,834 | bf16_mfu: 53.80 | total time: 43.89m | eta: 67.1m step 02209/05568 (39.67%) | loss: 2.603656 | lrm: 0.93 | dt: 1199.39ms | tok/sec: 874,256 | bf16_mfu: 53.70 | total time: 43.91m | eta: 67.1m step 02210/05568 (39.69%) | loss: 2.608134 | lrm: 0.93 | dt: 1194.72ms | tok/sec: 877,672 | bf16_mfu: 53.91 | total time: 43.93m | eta: 67.0m step 02211/05568 (39.71%) | loss: 2.608543 | lrm: 0.93 | dt: 1195.24ms | tok/sec: 877,291 | bf16_mfu: 53.89 | total time: 43.95m | eta: 67.0m step 02212/05568 (39.73%) | loss: 2.618830 | lrm: 0.93 | dt: 1202.39ms | tok/sec: 872,074 | bf16_mfu: 53.57 | total time: 43.97m | eta: 67.0m step 02213/05568 (39.74%) | loss: 2.625980 | lrm: 0.93 | dt: 1198.62ms | tok/sec: 874,821 | bf16_mfu: 53.74 | total time: 43.99m | eta: 67.0m step 02214/05568 (39.76%) | loss: 2.625965 | lrm: 0.93 | dt: 1198.06ms | tok/sec: 875,229 | bf16_mfu: 53.76 | total time: 44.01m | eta: 67.0m step 02215/05568 (39.78%) | loss: 2.627906 | lrm: 0.93 | dt: 1201.23ms | tok/sec: 872,918 | bf16_mfu: 53.62 | total time: 44.03m | eta: 66.9m step 02216/05568 (39.80%) | loss: 2.631779 | lrm: 0.93 | dt: 1201.80ms | tok/sec: 872,502 | bf16_mfu: 53.60 | total time: 44.05m | eta: 66.9m step 02217/05568 (39.82%) | loss: 2.630436 | lrm: 0.93 | dt: 1203.82ms | tok/sec: 871,044 | bf16_mfu: 53.51 | total time: 44.07m | eta: 66.9m step 02218/05568 (39.83%) | loss: 2.634062 | lrm: 0.93 | dt: 1194.40ms | tok/sec: 877,911 | bf16_mfu: 53.93 | total time: 44.09m | eta: 66.9m step 02219/05568 (39.85%) | loss: 2.637836 | lrm: 0.93 | dt: 1196.81ms | tok/sec: 876,141 | bf16_mfu: 53.82 | total time: 44.11m | eta: 66.9m step 02220/05568 (39.87%) | loss: 2.637701 | lrm: 0.93 | dt: 1202.99ms | tok/sec: 871,641 | bf16_mfu: 53.54 | total time: 44.13m | eta: 66.8m step 02221/05568 (39.89%) | loss: 2.643822 | lrm: 0.93 | dt: 1197.92ms | tok/sec: 875,334 | bf16_mfu: 53.77 | total time: 44.15m | eta: 66.8m step 02222/05568 (39.91%) | loss: 2.641696 | lrm: 0.93 | dt: 1198.82ms | tok/sec: 874,675 | bf16_mfu: 53.73 | total time: 44.17m | eta: 66.8m step 02223/05568 (39.92%) | loss: 2.637308 | lrm: 0.93 | dt: 1204.36ms | tok/sec: 870,649 | bf16_mfu: 53.48 | total time: 44.19m | eta: 66.8m step 02224/05568 (39.94%) | loss: 2.633940 | lrm: 0.93 | dt: 1200.24ms | tok/sec: 873,640 | bf16_mfu: 53.67 | total time: 44.21m | eta: 66.8m step 02225/05568 (39.96%) | loss: 2.636779 | lrm: 0.93 | dt: 1201.45ms | tok/sec: 872,760 | bf16_mfu: 53.61 | total time: 44.23m | eta: 66.7m step 02226/05568 (39.98%) | loss: 2.643760 | lrm: 0.93 | dt: 1196.76ms | tok/sec: 876,176 | bf16_mfu: 53.82 | total time: 44.25m | eta: 66.7m step 02227/05568 (40.00%) | loss: 2.637557 | lrm: 0.93 | dt: 1202.12ms | tok/sec: 872,269 | bf16_mfu: 53.58 | total time: 44.27m | eta: 66.7m step 02228/05568 (40.01%) | loss: 2.632612 | lrm: 0.93 | dt: 1197.13ms | tok/sec: 875,910 | bf16_mfu: 53.81 | total time: 44.29m | eta: 66.7m step 02229/05568 (40.03%) | loss: 2.644959 | lrm: 0.93 | dt: 1201.59ms | tok/sec: 872,654 | bf16_mfu: 53.61 | total time: 44.31m | eta: 66.7m step 02230/05568 (40.05%) | loss: 2.645306 | lrm: 0.93 | dt: 1201.51ms | tok/sec: 872,715 | bf16_mfu: 53.61 | total time: 44.33m | eta: 66.6m step 02231/05568 (40.07%) | loss: 2.639863 | lrm: 0.93 | dt: 1200.22ms | tok/sec: 873,652 | bf16_mfu: 53.67 | total time: 44.35m | eta: 66.6m step 02232/05568 (40.09%) | loss: 2.633138 | lrm: 0.93 | dt: 1198.37ms | tok/sec: 875,001 | bf16_mfu: 53.75 | total time: 44.37m | eta: 66.6m step 02233/05568 (40.10%) | loss: 2.634639 | lrm: 0.93 | dt: 1198.12ms | tok/sec: 875,186 | bf16_mfu: 53.76 | total time: 44.39m | eta: 66.6m step 02234/05568 (40.12%) | loss: 2.641562 | lrm: 0.93 | dt: 1198.21ms | tok/sec: 875,119 | bf16_mfu: 53.76 | total time: 44.41m | eta: 66.6m step 02235/05568 (40.14%) | loss: 2.631042 | lrm: 0.92 | dt: 1193.60ms | tok/sec: 878,495 | bf16_mfu: 53.96 | total time: 44.43m | eta: 66.5m step 02236/05568 (40.16%) | loss: 2.610519 | lrm: 0.92 | dt: 1197.72ms | tok/sec: 875,476 | bf16_mfu: 53.78 | total time: 44.45m | eta: 66.5m step 02237/05568 (40.18%) | loss: 2.610496 | lrm: 0.92 | dt: 1198.43ms | tok/sec: 874,961 | bf16_mfu: 53.75 | total time: 44.47m | eta: 66.5m step 02238/05568 (40.19%) | loss: 2.612339 | lrm: 0.92 | dt: 1195.62ms | tok/sec: 877,017 | bf16_mfu: 53.87 | total time: 44.49m | eta: 66.5m step 02239/05568 (40.21%) | loss: 2.615293 | lrm: 0.92 | dt: 1204.90ms | tok/sec: 870,257 | bf16_mfu: 53.46 | total time: 44.51m | eta: 66.5m step 02240/05568 (40.23%) | loss: 2.609013 | lrm: 0.92 | dt: 1197.37ms | tok/sec: 875,729 | bf16_mfu: 53.79 | total time: 44.53m | eta: 66.4m step 02241/05568 (40.25%) | loss: 2.613696 | lrm: 0.92 | dt: 1194.85ms | tok/sec: 877,579 | bf16_mfu: 53.91 | total time: 44.55m | eta: 66.4m step 02242/05568 (40.27%) | loss: 2.627337 | lrm: 0.92 | dt: 1197.84ms | tok/sec: 875,385 | bf16_mfu: 53.77 | total time: 44.57m | eta: 66.4m step 02243/05568 (40.28%) | loss: 2.626176 | lrm: 0.92 | dt: 1194.72ms | tok/sec: 877,674 | bf16_mfu: 53.91 | total time: 44.59m | eta: 66.4m step 02244/05568 (40.30%) | loss: 2.616150 | lrm: 0.92 | dt: 1197.99ms | tok/sec: 875,281 | bf16_mfu: 53.77 | total time: 44.61m | eta: 66.4m step 02245/05568 (40.32%) | loss: 2.616438 | lrm: 0.92 | dt: 1203.16ms | tok/sec: 871,516 | bf16_mfu: 53.54 | total time: 44.63m | eta: 66.3m step 02246/05568 (40.34%) | loss: 2.621353 | lrm: 0.92 | dt: 1196.46ms | tok/sec: 876,399 | bf16_mfu: 53.84 | total time: 44.65m | eta: 66.3m step 02247/05568 (40.36%) | loss: 2.635627 | lrm: 0.92 | dt: 1198.51ms | tok/sec: 874,902 | bf16_mfu: 53.74 | total time: 44.67m | eta: 66.3m step 02248/05568 (40.37%) | loss: 2.636229 | lrm: 0.92 | dt: 1199.80ms | tok/sec: 873,960 | bf16_mfu: 53.69 | total time: 44.69m | eta: 66.3m step 02249/05568 (40.39%) | loss: 2.631079 | lrm: 0.92 | dt: 1203.67ms | tok/sec: 871,147 | bf16_mfu: 53.51 | total time: 44.71m | eta: 66.3m step:2250/5568 val_bpb:0.798533 val_time:3.93s step 02250/05568 (40.41%) | loss: 2.631109 | lrm: 0.92 | dt: 1201.45ms | tok/sec: 872,760 | bf16_mfu: 53.61 | total time: 44.73m | eta: 66.2m step 02251/05568 (40.43%) | loss: 2.624107 | lrm: 0.92 | dt: 1215.73ms | tok/sec: 862,507 | bf16_mfu: 52.98 | total time: 44.75m | eta: 66.2m step 02252/05568 (40.45%) | loss: 2.624211 | lrm: 0.92 | dt: 1208.24ms | tok/sec: 867,857 | bf16_mfu: 53.31 | total time: 44.77m | eta: 66.2m step 02253/05568 (40.46%) | loss: 2.623117 | lrm: 0.92 | dt: 1204.33ms | tok/sec: 870,672 | bf16_mfu: 53.48 | total time: 44.79m | eta: 66.2m step 02254/05568 (40.48%) | loss: 2.621755 | lrm: 0.92 | dt: 1218.56ms | tok/sec: 860,505 | bf16_mfu: 52.86 | total time: 44.81m | eta: 66.2m step 02255/05568 (40.50%) | loss: 2.633290 | lrm: 0.92 | dt: 1203.80ms | tok/sec: 871,053 | bf16_mfu: 53.51 | total time: 44.83m | eta: 66.1m step 02256/05568 (40.52%) | loss: 2.628255 | lrm: 0.92 | dt: 1213.35ms | tok/sec: 864,196 | bf16_mfu: 53.09 | total time: 44.85m | eta: 66.1m step 02257/05568 (40.54%) | loss: 2.619223 | lrm: 0.92 | dt: 1210.50ms | tok/sec: 866,231 | bf16_mfu: 53.21 | total time: 44.87m | eta: 66.1m step 02258/05568 (40.55%) | loss: 2.630562 | lrm: 0.92 | dt: 1198.39ms | tok/sec: 874,987 | bf16_mfu: 53.75 | total time: 44.89m | eta: 66.1m step 02259/05568 (40.57%) | loss: 2.636013 | lrm: 0.92 | dt: 1197.98ms | tok/sec: 875,285 | bf16_mfu: 53.77 | total time: 44.91m | eta: 66.1m step 02260/05568 (40.59%) | loss: 2.633225 | lrm: 0.92 | dt: 1193.96ms | tok/sec: 878,234 | bf16_mfu: 53.95 | total time: 44.93m | eta: 66.0m step 02261/05568 (40.61%) | loss: 2.639782 | lrm: 0.92 | dt: 1197.83ms | tok/sec: 875,397 | bf16_mfu: 53.77 | total time: 44.95m | eta: 66.0m step 02262/05568 (40.62%) | loss: 2.645540 | lrm: 0.92 | dt: 1197.40ms | tok/sec: 875,712 | bf16_mfu: 53.79 | total time: 44.97m | eta: 66.0m step 02263/05568 (40.64%) | loss: 2.644297 | lrm: 0.92 | dt: 1192.47ms | tok/sec: 879,329 | bf16_mfu: 54.02 | total time: 44.99m | eta: 66.0m step 02264/05568 (40.66%) | loss: 2.643749 | lrm: 0.92 | dt: 1192.98ms | tok/sec: 878,953 | bf16_mfu: 53.99 | total time: 45.01m | eta: 66.0m step 02265/05568 (40.68%) | loss: 2.640041 | lrm: 0.92 | dt: 1201.17ms | tok/sec: 872,962 | bf16_mfu: 53.62 | total time: 45.03m | eta: 65.9m step 02266/05568 (40.70%) | loss: 2.638050 | lrm: 0.92 | dt: 1197.80ms | tok/sec: 875,415 | bf16_mfu: 53.77 | total time: 45.05m | eta: 65.9m step 02267/05568 (40.71%) | loss: 2.645948 | lrm: 0.92 | dt: 1188.85ms | tok/sec: 882,008 | bf16_mfu: 54.18 | total time: 45.07m | eta: 65.9m step 02268/05568 (40.73%) | loss: 2.649346 | lrm: 0.92 | dt: 1203.67ms | tok/sec: 871,147 | bf16_mfu: 53.51 | total time: 45.09m | eta: 65.9m step 02269/05568 (40.75%) | loss: 2.655502 | lrm: 0.92 | dt: 1196.18ms | tok/sec: 876,604 | bf16_mfu: 53.85 | total time: 45.11m | eta: 65.9m step 02270/05568 (40.77%) | loss: 2.647606 | lrm: 0.92 | dt: 1187.69ms | tok/sec: 882,867 | bf16_mfu: 54.23 | total time: 45.13m | eta: 65.8m step 02271/05568 (40.79%) | loss: 2.641360 | lrm: 0.92 | dt: 1195.79ms | tok/sec: 876,888 | bf16_mfu: 53.87 | total time: 45.15m | eta: 65.8m step 02272/05568 (40.80%) | loss: 2.631315 | lrm: 0.92 | dt: 1195.39ms | tok/sec: 877,184 | bf16_mfu: 53.88 | total time: 45.17m | eta: 65.8m step 02273/05568 (40.82%) | loss: 2.640304 | lrm: 0.91 | dt: 1193.55ms | tok/sec: 878,535 | bf16_mfu: 53.97 | total time: 45.19m | eta: 65.8m step 02274/05568 (40.84%) | loss: 2.640061 | lrm: 0.91 | dt: 1190.35ms | tok/sec: 880,896 | bf16_mfu: 54.11 | total time: 45.21m | eta: 65.8m step 02275/05568 (40.86%) | loss: 2.632483 | lrm: 0.91 | dt: 1191.33ms | tok/sec: 880,172 | bf16_mfu: 54.07 | total time: 45.23m | eta: 65.7m step 02276/05568 (40.88%) | loss: 2.629406 | lrm: 0.91 | dt: 1194.68ms | tok/sec: 877,703 | bf16_mfu: 53.92 | total time: 45.25m | eta: 65.7m step 02277/05568 (40.89%) | loss: 2.627044 | lrm: 0.91 | dt: 1191.86ms | tok/sec: 879,781 | bf16_mfu: 54.04 | total time: 45.27m | eta: 65.7m step 02278/05568 (40.91%) | loss: 2.624710 | lrm: 0.91 | dt: 1196.70ms | tok/sec: 876,226 | bf16_mfu: 53.82 | total time: 45.29m | eta: 65.7m step 02279/05568 (40.93%) | loss: 2.623079 | lrm: 0.91 | dt: 1191.54ms | tok/sec: 880,019 | bf16_mfu: 54.06 | total time: 45.31m | eta: 65.7m step 02280/05568 (40.95%) | loss: 2.613121 | lrm: 0.91 | dt: 1199.01ms | tok/sec: 874,535 | bf16_mfu: 53.72 | total time: 45.33m | eta: 65.6m step 02281/05568 (40.97%) | loss: 2.619380 | lrm: 0.91 | dt: 1194.23ms | tok/sec: 878,033 | bf16_mfu: 53.94 | total time: 45.35m | eta: 65.6m step 02282/05568 (40.98%) | loss: 2.619409 | lrm: 0.91 | dt: 1189.22ms | tok/sec: 881,735 | bf16_mfu: 54.16 | total time: 45.37m | eta: 65.6m step 02283/05568 (41.00%) | loss: 2.610486 | lrm: 0.91 | dt: 1189.14ms | tok/sec: 881,796 | bf16_mfu: 54.17 | total time: 45.39m | eta: 65.6m step 02284/05568 (41.02%) | loss: 2.613161 | lrm: 0.91 | dt: 1188.55ms | tok/sec: 882,228 | bf16_mfu: 54.19 | total time: 45.41m | eta: 65.6m step 02285/05568 (41.04%) | loss: 2.619615 | lrm: 0.91 | dt: 1192.83ms | tok/sec: 879,066 | bf16_mfu: 54.00 | total time: 45.43m | eta: 65.5m step 02286/05568 (41.06%) | loss: 2.615431 | lrm: 0.91 | dt: 1192.64ms | tok/sec: 879,205 | bf16_mfu: 54.01 | total time: 45.45m | eta: 65.5m step 02287/05568 (41.07%) | loss: 2.633401 | lrm: 0.91 | dt: 1198.23ms | tok/sec: 875,104 | bf16_mfu: 53.76 | total time: 45.47m | eta: 65.5m step 02288/05568 (41.09%) | loss: 2.634160 | lrm: 0.91 | dt: 1188.49ms | tok/sec: 882,273 | bf16_mfu: 54.20 | total time: 45.49m | eta: 65.5m step 02289/05568 (41.11%) | loss: 2.632874 | lrm: 0.91 | dt: 1195.60ms | tok/sec: 877,031 | bf16_mfu: 53.87 | total time: 45.51m | eta: 65.5m step 02290/05568 (41.13%) | loss: 2.641328 | lrm: 0.91 | dt: 1190.59ms | tok/sec: 880,718 | bf16_mfu: 54.10 | total time: 45.53m | eta: 65.4m step 02291/05568 (41.15%) | loss: 2.654487 | lrm: 0.91 | dt: 1195.31ms | tok/sec: 877,244 | bf16_mfu: 53.89 | total time: 45.55m | eta: 65.4m step 02292/05568 (41.16%) | loss: 2.654748 | lrm: 0.91 | dt: 1191.50ms | tok/sec: 880,048 | bf16_mfu: 54.06 | total time: 45.57m | eta: 65.4m step 02293/05568 (41.18%) | loss: 2.656052 | lrm: 0.91 | dt: 1190.40ms | tok/sec: 880,858 | bf16_mfu: 54.11 | total time: 45.59m | eta: 65.4m step 02294/05568 (41.20%) | loss: 2.650943 | lrm: 0.91 | dt: 1191.40ms | tok/sec: 880,123 | bf16_mfu: 54.06 | total time: 45.61m | eta: 65.4m step 02295/05568 (41.22%) | loss: 2.654124 | lrm: 0.91 | dt: 1193.64ms | tok/sec: 878,467 | bf16_mfu: 53.96 | total time: 45.63m | eta: 65.3m step 02296/05568 (41.24%) | loss: 2.661812 | lrm: 0.91 | dt: 1190.41ms | tok/sec: 880,853 | bf16_mfu: 54.11 | total time: 45.65m | eta: 65.3m step 02297/05568 (41.25%) | loss: 2.656561 | lrm: 0.91 | dt: 1197.85ms | tok/sec: 875,379 | bf16_mfu: 53.77 | total time: 45.67m | eta: 65.3m step 02298/05568 (41.27%) | loss: 2.665118 | lrm: 0.91 | dt: 1194.91ms | tok/sec: 877,538 | bf16_mfu: 53.91 | total time: 45.69m | eta: 65.3m step 02299/05568 (41.29%) | loss: 2.655374 | lrm: 0.91 | dt: 1191.20ms | tok/sec: 880,270 | bf16_mfu: 54.07 | total time: 45.71m | eta: 65.3m step 02300/05568 (41.31%) | loss: 2.657114 | lrm: 0.91 | dt: 1193.10ms | tok/sec: 878,866 | bf16_mfu: 53.99 | total time: 45.73m | eta: 65.2m step 02301/05568 (41.33%) | loss: 2.651807 | lrm: 0.91 | dt: 1186.12ms | tok/sec: 884,035 | bf16_mfu: 54.30 | total time: 45.75m | eta: 65.2m step 02302/05568 (41.34%) | loss: 2.649817 | lrm: 0.91 | dt: 1195.70ms | tok/sec: 876,955 | bf16_mfu: 53.87 | total time: 45.77m | eta: 65.2m step 02303/05568 (41.36%) | loss: 2.650984 | lrm: 0.91 | dt: 1190.66ms | tok/sec: 880,665 | bf16_mfu: 54.10 | total time: 45.79m | eta: 65.2m step 02304/05568 (41.38%) | loss: 2.646704 | lrm: 0.91 | dt: 1198.94ms | tok/sec: 874,583 | bf16_mfu: 53.72 | total time: 45.81m | eta: 65.2m step 02305/05568 (41.40%) | loss: 2.650892 | lrm: 0.91 | dt: 1198.83ms | tok/sec: 874,665 | bf16_mfu: 53.73 | total time: 45.83m | eta: 65.1m step 02306/05568 (41.42%) | loss: 2.654944 | lrm: 0.91 | dt: 1190.28ms | tok/sec: 880,951 | bf16_mfu: 54.11 | total time: 45.85m | eta: 65.1m step 02307/05568 (41.43%) | loss: 2.645108 | lrm: 0.91 | dt: 1196.54ms | tok/sec: 876,340 | bf16_mfu: 53.83 | total time: 45.87m | eta: 65.1m step 02308/05568 (41.45%) | loss: 2.659208 | lrm: 0.91 | dt: 1187.01ms | tok/sec: 883,375 | bf16_mfu: 54.26 | total time: 45.89m | eta: 65.1m step 02309/05568 (41.47%) | loss: 2.659518 | lrm: 0.91 | dt: 1187.13ms | tok/sec: 883,283 | bf16_mfu: 54.26 | total time: 45.91m | eta: 65.1m step 02310/05568 (41.49%) | loss: 2.661057 | lrm: 0.91 | dt: 1192.98ms | tok/sec: 878,953 | bf16_mfu: 53.99 | total time: 45.93m | eta: 65.0m step 02311/05568 (41.51%) | loss: 2.657832 | lrm: 0.90 | dt: 1190.48ms | tok/sec: 880,799 | bf16_mfu: 54.11 | total time: 45.95m | eta: 65.0m step 02312/05568 (41.52%) | loss: 2.655417 | lrm: 0.90 | dt: 1194.18ms | tok/sec: 878,069 | bf16_mfu: 53.94 | total time: 45.97m | eta: 65.0m step 02313/05568 (41.54%) | loss: 2.666543 | lrm: 0.90 | dt: 1194.43ms | tok/sec: 877,889 | bf16_mfu: 53.93 | total time: 45.99m | eta: 65.0m step 02314/05568 (41.56%) | loss: 2.655830 | lrm: 0.90 | dt: 1198.06ms | tok/sec: 875,226 | bf16_mfu: 53.76 | total time: 46.01m | eta: 65.0m step 02315/05568 (41.58%) | loss: 2.649863 | lrm: 0.90 | dt: 1196.89ms | tok/sec: 876,082 | bf16_mfu: 53.82 | total time: 46.03m | eta: 64.9m step 02316/05568 (41.59%) | loss: 2.656722 | lrm: 0.90 | dt: 1193.92ms | tok/sec: 878,265 | bf16_mfu: 53.95 | total time: 46.05m | eta: 64.9m step 02317/05568 (41.61%) | loss: 2.666957 | lrm: 0.90 | dt: 1193.89ms | tok/sec: 878,287 | bf16_mfu: 53.95 | total time: 46.07m | eta: 64.9m step 02318/05568 (41.63%) | loss: 2.657930 | lrm: 0.90 | dt: 1195.03ms | tok/sec: 877,443 | bf16_mfu: 53.90 | total time: 46.09m | eta: 64.9m step 02319/05568 (41.65%) | loss: 2.638085 | lrm: 0.90 | dt: 1190.66ms | tok/sec: 880,665 | bf16_mfu: 54.10 | total time: 46.11m | eta: 64.9m step 02320/05568 (41.67%) | loss: 2.637943 | lrm: 0.90 | dt: 1191.79ms | tok/sec: 879,830 | bf16_mfu: 54.05 | total time: 46.13m | eta: 64.8m step 02321/05568 (41.68%) | loss: 2.641431 | lrm: 0.90 | dt: 1192.67ms | tok/sec: 879,186 | bf16_mfu: 54.01 | total time: 46.15m | eta: 64.8m step 02322/05568 (41.70%) | loss: 2.637860 | lrm: 0.90 | dt: 1195.76ms | tok/sec: 876,908 | bf16_mfu: 53.87 | total time: 46.17m | eta: 64.8m step 02323/05568 (41.72%) | loss: 2.645777 | lrm: 0.90 | dt: 1193.92ms | tok/sec: 878,266 | bf16_mfu: 53.95 | total time: 46.19m | eta: 64.8m step 02324/05568 (41.74%) | loss: 2.648748 | lrm: 0.90 | dt: 1189.98ms | tok/sec: 881,169 | bf16_mfu: 54.13 | total time: 46.21m | eta: 64.8m step 02325/05568 (41.76%) | loss: 2.647021 | lrm: 0.90 | dt: 1189.71ms | tok/sec: 881,372 | bf16_mfu: 54.14 | total time: 46.23m | eta: 64.7m step 02326/05568 (41.77%) | loss: 2.653244 | lrm: 0.90 | dt: 1191.47ms | tok/sec: 880,065 | bf16_mfu: 54.06 | total time: 46.25m | eta: 64.7m step 02327/05568 (41.79%) | loss: 2.642091 | lrm: 0.90 | dt: 1191.90ms | tok/sec: 879,751 | bf16_mfu: 54.04 | total time: 46.27m | eta: 64.7m step 02328/05568 (41.81%) | loss: 2.634157 | lrm: 0.90 | dt: 1196.96ms | tok/sec: 876,031 | bf16_mfu: 53.81 | total time: 46.29m | eta: 64.7m step 02329/05568 (41.83%) | loss: 2.628472 | lrm: 0.90 | dt: 1193.16ms | tok/sec: 878,824 | bf16_mfu: 53.98 | total time: 46.31m | eta: 64.7m step 02330/05568 (41.85%) | loss: 2.630180 | lrm: 0.90 | dt: 1193.12ms | tok/sec: 878,855 | bf16_mfu: 53.99 | total time: 46.33m | eta: 64.6m step 02331/05568 (41.86%) | loss: 2.632859 | lrm: 0.90 | dt: 1194.10ms | tok/sec: 878,130 | bf16_mfu: 53.94 | total time: 46.35m | eta: 64.6m step 02332/05568 (41.88%) | loss: 2.614387 | lrm: 0.90 | dt: 1195.88ms | tok/sec: 876,825 | bf16_mfu: 53.86 | total time: 46.37m | eta: 64.6m step 02333/05568 (41.90%) | loss: 2.618531 | lrm: 0.90 | dt: 1191.16ms | tok/sec: 880,299 | bf16_mfu: 54.07 | total time: 46.39m | eta: 64.6m step 02334/05568 (41.92%) | loss: 2.621096 | lrm: 0.90 | dt: 1193.21ms | tok/sec: 878,783 | bf16_mfu: 53.98 | total time: 46.41m | eta: 64.6m step 02335/05568 (41.94%) | loss: 2.630355 | lrm: 0.90 | dt: 1194.90ms | tok/sec: 877,542 | bf16_mfu: 53.91 | total time: 46.43m | eta: 64.5m step 02336/05568 (41.95%) | loss: 2.633807 | lrm: 0.90 | dt: 1194.41ms | tok/sec: 877,900 | bf16_mfu: 53.93 | total time: 46.45m | eta: 64.5m step 02337/05568 (41.97%) | loss: 2.634630 | lrm: 0.90 | dt: 1190.60ms | tok/sec: 880,711 | bf16_mfu: 54.10 | total time: 46.47m | eta: 64.5m step 02338/05568 (41.99%) | loss: 2.635030 | lrm: 0.90 | dt: 1190.60ms | tok/sec: 880,710 | bf16_mfu: 54.10 | total time: 46.49m | eta: 64.5m step 02339/05568 (42.01%) | loss: 2.631674 | lrm: 0.90 | dt: 1189.59ms | tok/sec: 881,459 | bf16_mfu: 54.15 | total time: 46.50m | eta: 64.5m step 02340/05568 (42.03%) | loss: 2.633921 | lrm: 0.90 | dt: 1190.68ms | tok/sec: 880,651 | bf16_mfu: 54.10 | total time: 46.52m | eta: 64.4m step 02341/05568 (42.04%) | loss: 2.633699 | lrm: 0.90 | dt: 1193.20ms | tok/sec: 878,796 | bf16_mfu: 53.98 | total time: 46.54m | eta: 64.4m step 02342/05568 (42.06%) | loss: 2.628424 | lrm: 0.90 | dt: 1187.01ms | tok/sec: 883,375 | bf16_mfu: 54.26 | total time: 46.56m | eta: 64.4m step 02343/05568 (42.08%) | loss: 2.632284 | lrm: 0.90 | dt: 1198.55ms | tok/sec: 874,868 | bf16_mfu: 53.74 | total time: 46.58m | eta: 64.4m step 02344/05568 (42.10%) | loss: 2.624020 | lrm: 0.90 | dt: 1192.86ms | tok/sec: 879,045 | bf16_mfu: 54.00 | total time: 46.60m | eta: 64.4m step 02345/05568 (42.12%) | loss: 2.625827 | lrm: 0.90 | dt: 1188.66ms | tok/sec: 882,150 | bf16_mfu: 54.19 | total time: 46.62m | eta: 64.3m step 02346/05568 (42.13%) | loss: 2.629405 | lrm: 0.90 | dt: 1191.76ms | tok/sec: 879,852 | bf16_mfu: 54.05 | total time: 46.64m | eta: 64.3m step 02347/05568 (42.15%) | loss: 2.634388 | lrm: 0.90 | dt: 1192.05ms | tok/sec: 879,637 | bf16_mfu: 54.03 | total time: 46.66m | eta: 64.3m step 02348/05568 (42.17%) | loss: 2.635333 | lrm: 0.90 | dt: 1188.09ms | tok/sec: 882,569 | bf16_mfu: 54.21 | total time: 46.68m | eta: 64.3m step 02349/05568 (42.19%) | loss: 2.630726 | lrm: 0.89 | dt: 1193.33ms | tok/sec: 878,700 | bf16_mfu: 53.98 | total time: 46.70m | eta: 64.3m step 02350/05568 (42.21%) | loss: 2.633031 | lrm: 0.89 | dt: 1189.25ms | tok/sec: 881,711 | bf16_mfu: 54.16 | total time: 46.72m | eta: 64.2m step 02351/05568 (42.22%) | loss: 2.625634 | lrm: 0.89 | dt: 1193.47ms | tok/sec: 878,593 | bf16_mfu: 53.97 | total time: 46.74m | eta: 64.2m step 02352/05568 (42.24%) | loss: 2.622557 | lrm: 0.89 | dt: 1194.97ms | tok/sec: 877,491 | bf16_mfu: 53.90 | total time: 46.76m | eta: 64.2m step 02353/05568 (42.26%) | loss: 2.629780 | lrm: 0.89 | dt: 1192.87ms | tok/sec: 879,037 | bf16_mfu: 54.00 | total time: 46.78m | eta: 64.2m step 02354/05568 (42.28%) | loss: 2.636677 | lrm: 0.89 | dt: 1195.16ms | tok/sec: 877,355 | bf16_mfu: 53.89 | total time: 46.80m | eta: 64.2m step 02355/05568 (42.30%) | loss: 2.630640 | lrm: 0.89 | dt: 1189.77ms | tok/sec: 881,325 | bf16_mfu: 54.14 | total time: 46.82m | eta: 64.1m step 02356/05568 (42.31%) | loss: 2.627125 | lrm: 0.89 | dt: 1194.12ms | tok/sec: 878,115 | bf16_mfu: 53.94 | total time: 46.84m | eta: 64.1m step 02357/05568 (42.33%) | loss: 2.626576 | lrm: 0.89 | dt: 1190.91ms | tok/sec: 880,485 | bf16_mfu: 54.09 | total time: 46.86m | eta: 64.1m step 02358/05568 (42.35%) | loss: 2.628469 | lrm: 0.89 | dt: 1193.64ms | tok/sec: 878,468 | bf16_mfu: 53.96 | total time: 46.88m | eta: 64.1m step 02359/05568 (42.37%) | loss: 2.634396 | lrm: 0.89 | dt: 1194.32ms | tok/sec: 877,966 | bf16_mfu: 53.93 | total time: 46.90m | eta: 64.1m step 02360/05568 (42.39%) | loss: 2.622262 | lrm: 0.89 | dt: 1190.89ms | tok/sec: 880,498 | bf16_mfu: 54.09 | total time: 46.92m | eta: 64.0m step 02361/05568 (42.40%) | loss: 2.630928 | lrm: 0.89 | dt: 1192.17ms | tok/sec: 879,552 | bf16_mfu: 54.03 | total time: 46.94m | eta: 64.0m step 02362/05568 (42.42%) | loss: 2.630935 | lrm: 0.89 | dt: 1191.87ms | tok/sec: 879,772 | bf16_mfu: 54.04 | total time: 46.96m | eta: 64.0m step 02363/05568 (42.44%) | loss: 2.628453 | lrm: 0.89 | dt: 1190.95ms | tok/sec: 880,450 | bf16_mfu: 54.08 | total time: 46.98m | eta: 64.0m step 02364/05568 (42.46%) | loss: 2.629005 | lrm: 0.89 | dt: 1192.01ms | tok/sec: 879,672 | bf16_mfu: 54.04 | total time: 47.00m | eta: 64.0m step 02365/05568 (42.47%) | loss: 2.638195 | lrm: 0.89 | dt: 1191.23ms | tok/sec: 880,246 | bf16_mfu: 54.07 | total time: 47.02m | eta: 63.9m step 02366/05568 (42.49%) | loss: 2.636374 | lrm: 0.89 | dt: 1191.16ms | tok/sec: 880,299 | bf16_mfu: 54.07 | total time: 47.04m | eta: 63.9m step 02367/05568 (42.51%) | loss: 2.640105 | lrm: 0.89 | dt: 1191.38ms | tok/sec: 880,132 | bf16_mfu: 54.06 | total time: 47.06m | eta: 63.9m step 02368/05568 (42.53%) | loss: 2.637564 | lrm: 0.89 | dt: 1192.06ms | tok/sec: 879,631 | bf16_mfu: 54.03 | total time: 47.08m | eta: 63.9m step 02369/05568 (42.55%) | loss: 2.641705 | lrm: 0.89 | dt: 1192.49ms | tok/sec: 879,319 | bf16_mfu: 54.01 | total time: 47.10m | eta: 63.9m step 02370/05568 (42.56%) | loss: 2.631207 | lrm: 0.89 | dt: 1192.39ms | tok/sec: 879,393 | bf16_mfu: 54.02 | total time: 47.12m | eta: 63.8m step 02371/05568 (42.58%) | loss: 2.635276 | lrm: 0.89 | dt: 1195.86ms | tok/sec: 876,841 | bf16_mfu: 53.86 | total time: 47.14m | eta: 63.8m step 02372/05568 (42.60%) | loss: 2.645441 | lrm: 0.89 | dt: 1193.76ms | tok/sec: 878,382 | bf16_mfu: 53.96 | total time: 47.16m | eta: 63.8m step 02373/05568 (42.62%) | loss: 2.628708 | lrm: 0.89 | dt: 1193.73ms | tok/sec: 878,402 | bf16_mfu: 53.96 | total time: 47.18m | eta: 63.8m step 02374/05568 (42.64%) | loss: 2.632111 | lrm: 0.89 | dt: 1190.93ms | tok/sec: 880,470 | bf16_mfu: 54.09 | total time: 47.20m | eta: 63.8m step 02375/05568 (42.65%) | loss: 2.645027 | lrm: 0.89 | dt: 1190.29ms | tok/sec: 880,944 | bf16_mfu: 54.11 | total time: 47.22m | eta: 63.7m step 02376/05568 (42.67%) | loss: 2.635446 | lrm: 0.89 | dt: 1194.37ms | tok/sec: 877,932 | bf16_mfu: 53.93 | total time: 47.24m | eta: 63.7m step 02377/05568 (42.69%) | loss: 2.636564 | lrm: 0.89 | dt: 1193.07ms | tok/sec: 878,891 | bf16_mfu: 53.99 | total time: 47.26m | eta: 63.7m step 02378/05568 (42.71%) | loss: 2.636273 | lrm: 0.89 | dt: 1190.14ms | tok/sec: 881,051 | bf16_mfu: 54.12 | total time: 47.28m | eta: 63.7m step 02379/05568 (42.73%) | loss: 2.633654 | lrm: 0.89 | dt: 1193.22ms | tok/sec: 878,778 | bf16_mfu: 53.98 | total time: 47.30m | eta: 63.7m step 02380/05568 (42.74%) | loss: 2.634730 | lrm: 0.89 | dt: 1192.86ms | tok/sec: 879,045 | bf16_mfu: 54.00 | total time: 47.32m | eta: 63.6m step 02381/05568 (42.76%) | loss: 2.637058 | lrm: 0.89 | dt: 1196.63ms | tok/sec: 876,276 | bf16_mfu: 53.83 | total time: 47.34m | eta: 63.6m step 02382/05568 (42.78%) | loss: 2.643218 | lrm: 0.89 | dt: 1189.96ms | tok/sec: 881,185 | bf16_mfu: 54.13 | total time: 47.36m | eta: 63.6m step 02383/05568 (42.80%) | loss: 2.644662 | lrm: 0.89 | dt: 1191.28ms | tok/sec: 880,210 | bf16_mfu: 54.07 | total time: 47.38m | eta: 63.6m step 02384/05568 (42.82%) | loss: 2.636335 | lrm: 0.89 | dt: 1190.76ms | tok/sec: 880,597 | bf16_mfu: 54.09 | total time: 47.40m | eta: 63.6m step 02385/05568 (42.83%) | loss: 2.639504 | lrm: 0.89 | dt: 1193.33ms | tok/sec: 878,694 | bf16_mfu: 53.98 | total time: 47.42m | eta: 63.5m step 02386/05568 (42.85%) | loss: 2.644227 | lrm: 0.89 | dt: 1191.52ms | tok/sec: 880,029 | bf16_mfu: 54.06 | total time: 47.44m | eta: 63.5m step 02387/05568 (42.87%) | loss: 2.648667 | lrm: 0.89 | dt: 1191.80ms | tok/sec: 879,822 | bf16_mfu: 54.05 | total time: 47.46m | eta: 63.5m step 02388/05568 (42.89%) | loss: 2.652305 | lrm: 0.88 | dt: 1191.23ms | tok/sec: 880,243 | bf16_mfu: 54.07 | total time: 47.48m | eta: 63.5m step 02389/05568 (42.91%) | loss: 2.642881 | lrm: 0.88 | dt: 1190.26ms | tok/sec: 880,965 | bf16_mfu: 54.12 | total time: 47.50m | eta: 63.5m step 02390/05568 (42.92%) | loss: 2.628595 | lrm: 0.88 | dt: 1197.07ms | tok/sec: 875,953 | bf16_mfu: 53.81 | total time: 47.52m | eta: 63.4m step 02391/05568 (42.94%) | loss: 2.615990 | lrm: 0.88 | dt: 1192.61ms | tok/sec: 879,227 | bf16_mfu: 54.01 | total time: 47.54m | eta: 63.4m step 02392/05568 (42.96%) | loss: 2.611346 | lrm: 0.88 | dt: 1190.96ms | tok/sec: 880,445 | bf16_mfu: 54.08 | total time: 47.56m | eta: 63.4m step 02393/05568 (42.98%) | loss: 2.604079 | lrm: 0.88 | dt: 1195.50ms | tok/sec: 877,099 | bf16_mfu: 53.88 | total time: 47.58m | eta: 63.4m step 02394/05568 (43.00%) | loss: 2.599484 | lrm: 0.88 | dt: 1193.85ms | tok/sec: 878,312 | bf16_mfu: 53.95 | total time: 47.60m | eta: 63.4m step 02395/05568 (43.01%) | loss: 2.592997 | lrm: 0.88 | dt: 1193.58ms | tok/sec: 878,515 | bf16_mfu: 53.97 | total time: 47.62m | eta: 63.3m step 02396/05568 (43.03%) | loss: 2.590650 | lrm: 0.88 | dt: 1192.38ms | tok/sec: 879,394 | bf16_mfu: 54.02 | total time: 47.64m | eta: 63.3m step 02397/05568 (43.05%) | loss: 2.592877 | lrm: 0.88 | dt: 1192.37ms | tok/sec: 879,402 | bf16_mfu: 54.02 | total time: 47.66m | eta: 63.3m step 02398/05568 (43.07%) | loss: 2.605738 | lrm: 0.88 | dt: 1189.11ms | tok/sec: 881,816 | bf16_mfu: 54.17 | total time: 47.68m | eta: 63.3m step 02399/05568 (43.09%) | loss: 2.610591 | lrm: 0.88 | dt: 1195.55ms | tok/sec: 877,068 | bf16_mfu: 53.88 | total time: 47.70m | eta: 63.3m step 02400/05568 (43.10%) | loss: 2.601978 | lrm: 0.88 | dt: 1194.85ms | tok/sec: 877,582 | bf16_mfu: 53.91 | total time: 47.72m | eta: 63.2m step 02401/05568 (43.12%) | loss: 2.600155 | lrm: 0.88 | dt: 1190.21ms | tok/sec: 880,997 | bf16_mfu: 54.12 | total time: 47.74m | eta: 63.2m step 02402/05568 (43.14%) | loss: 2.586766 | lrm: 0.88 | dt: 1196.91ms | tok/sec: 876,069 | bf16_mfu: 53.81 | total time: 47.76m | eta: 63.2m step 02403/05568 (43.16%) | loss: 2.579578 | lrm: 0.88 | dt: 1193.76ms | tok/sec: 878,381 | bf16_mfu: 53.96 | total time: 47.78m | eta: 63.2m step 02404/05568 (43.18%) | loss: 2.580048 | lrm: 0.88 | dt: 1194.25ms | tok/sec: 878,018 | bf16_mfu: 53.93 | total time: 47.80m | eta: 63.1m step 02405/05568 (43.19%) | loss: 2.579059 | lrm: 0.88 | dt: 1189.87ms | tok/sec: 881,251 | bf16_mfu: 54.13 | total time: 47.82m | eta: 63.1m step 02406/05568 (43.21%) | loss: 2.586211 | lrm: 0.88 | dt: 1194.73ms | tok/sec: 877,664 | bf16_mfu: 53.91 | total time: 47.84m | eta: 63.1m step 02407/05568 (43.23%) | loss: 2.596150 | lrm: 0.88 | dt: 1189.44ms | tok/sec: 881,570 | bf16_mfu: 54.15 | total time: 47.86m | eta: 63.1m step 02408/05568 (43.25%) | loss: 2.596834 | lrm: 0.88 | dt: 1190.94ms | tok/sec: 880,458 | bf16_mfu: 54.08 | total time: 47.88m | eta: 63.1m step 02409/05568 (43.27%) | loss: 2.607699 | lrm: 0.88 | dt: 1188.86ms | tok/sec: 881,998 | bf16_mfu: 54.18 | total time: 47.90m | eta: 63.0m step 02410/05568 (43.28%) | loss: 2.598799 | lrm: 0.88 | dt: 1193.19ms | tok/sec: 878,800 | bf16_mfu: 53.98 | total time: 47.92m | eta: 63.0m step 02411/05568 (43.30%) | loss: 2.600112 | lrm: 0.88 | dt: 1191.74ms | tok/sec: 879,869 | bf16_mfu: 54.05 | total time: 47.94m | eta: 63.0m step 02412/05568 (43.32%) | loss: 2.588747 | lrm: 0.88 | dt: 1191.99ms | tok/sec: 879,684 | bf16_mfu: 54.04 | total time: 47.96m | eta: 63.0m step 02413/05568 (43.34%) | loss: 2.598781 | lrm: 0.88 | dt: 1193.04ms | tok/sec: 878,911 | bf16_mfu: 53.99 | total time: 47.98m | eta: 63.0m step 02414/05568 (43.35%) | loss: 2.610138 | lrm: 0.88 | dt: 1191.70ms | tok/sec: 879,896 | bf16_mfu: 54.05 | total time: 48.00m | eta: 62.9m step 02415/05568 (43.37%) | loss: 2.611415 | lrm: 0.88 | dt: 1189.47ms | tok/sec: 881,551 | bf16_mfu: 54.15 | total time: 48.02m | eta: 62.9m step 02416/05568 (43.39%) | loss: 2.613388 | lrm: 0.88 | dt: 1193.29ms | tok/sec: 878,727 | bf16_mfu: 53.98 | total time: 48.04m | eta: 62.9m step 02417/05568 (43.41%) | loss: 2.612788 | lrm: 0.88 | dt: 1189.56ms | tok/sec: 881,482 | bf16_mfu: 54.15 | total time: 48.05m | eta: 62.9m step 02418/05568 (43.43%) | loss: 2.615343 | lrm: 0.88 | dt: 1190.55ms | tok/sec: 880,748 | bf16_mfu: 54.10 | total time: 48.07m | eta: 62.9m step 02419/05568 (43.44%) | loss: 2.612888 | lrm: 0.88 | dt: 1191.91ms | tok/sec: 879,747 | bf16_mfu: 54.04 | total time: 48.09m | eta: 62.8m step 02420/05568 (43.46%) | loss: 2.618814 | lrm: 0.88 | dt: 1193.13ms | tok/sec: 878,845 | bf16_mfu: 53.99 | total time: 48.11m | eta: 62.8m step 02421/05568 (43.48%) | loss: 2.617303 | lrm: 0.88 | dt: 1186.83ms | tok/sec: 883,507 | bf16_mfu: 54.27 | total time: 48.13m | eta: 62.8m step 02422/05568 (43.50%) | loss: 2.604345 | lrm: 0.88 | dt: 1192.42ms | tok/sec: 879,369 | bf16_mfu: 54.02 | total time: 48.15m | eta: 62.8m step 02423/05568 (43.52%) | loss: 2.616616 | lrm: 0.88 | dt: 1191.00ms | tok/sec: 880,413 | bf16_mfu: 54.08 | total time: 48.17m | eta: 62.8m step 02424/05568 (43.53%) | loss: 2.610471 | lrm: 0.88 | dt: 1190.98ms | tok/sec: 880,428 | bf16_mfu: 54.08 | total time: 48.19m | eta: 62.7m step 02425/05568 (43.55%) | loss: 2.606946 | lrm: 0.88 | dt: 1189.98ms | tok/sec: 881,173 | bf16_mfu: 54.13 | total time: 48.21m | eta: 62.7m step 02426/05568 (43.57%) | loss: 2.601949 | lrm: 0.87 | dt: 1190.31ms | tok/sec: 880,928 | bf16_mfu: 54.11 | total time: 48.23m | eta: 62.7m step 02427/05568 (43.59%) | loss: 2.597493 | lrm: 0.87 | dt: 1194.48ms | tok/sec: 877,851 | bf16_mfu: 53.92 | total time: 48.25m | eta: 62.7m step 02428/05568 (43.61%) | loss: 2.591838 | lrm: 0.87 | dt: 1189.00ms | tok/sec: 881,901 | bf16_mfu: 54.17 | total time: 48.27m | eta: 62.7m step 02429/05568 (43.62%) | loss: 2.587146 | lrm: 0.87 | dt: 1193.98ms | tok/sec: 878,222 | bf16_mfu: 53.95 | total time: 48.29m | eta: 62.6m step 02430/05568 (43.64%) | loss: 2.580041 | lrm: 0.87 | dt: 1193.56ms | tok/sec: 878,529 | bf16_mfu: 53.97 | total time: 48.31m | eta: 62.6m step 02431/05568 (43.66%) | loss: 2.589955 | lrm: 0.87 | dt: 1190.51ms | tok/sec: 880,779 | bf16_mfu: 54.10 | total time: 48.33m | eta: 62.6m step 02432/05568 (43.68%) | loss: 2.589894 | lrm: 0.87 | dt: 1190.85ms | tok/sec: 880,530 | bf16_mfu: 54.09 | total time: 48.35m | eta: 62.6m step 02433/05568 (43.70%) | loss: 2.591440 | lrm: 0.87 | dt: 1191.25ms | tok/sec: 880,233 | bf16_mfu: 54.07 | total time: 48.37m | eta: 62.6m step 02434/05568 (43.71%) | loss: 2.587234 | lrm: 0.87 | dt: 1194.64ms | tok/sec: 877,735 | bf16_mfu: 53.92 | total time: 48.39m | eta: 62.5m step 02435/05568 (43.73%) | loss: 2.587066 | lrm: 0.87 | dt: 1193.55ms | tok/sec: 878,537 | bf16_mfu: 53.97 | total time: 48.41m | eta: 62.5m step 02436/05568 (43.75%) | loss: 2.604382 | lrm: 0.87 | dt: 1192.40ms | tok/sec: 879,379 | bf16_mfu: 54.02 | total time: 48.43m | eta: 62.5m step 02437/05568 (43.77%) | loss: 2.606259 | lrm: 0.87 | dt: 1190.68ms | tok/sec: 880,649 | bf16_mfu: 54.10 | total time: 48.45m | eta: 62.5m step 02438/05568 (43.79%) | loss: 2.610494 | lrm: 0.87 | dt: 1192.01ms | tok/sec: 879,673 | bf16_mfu: 54.04 | total time: 48.47m | eta: 62.5m step 02439/05568 (43.80%) | loss: 2.608741 | lrm: 0.87 | dt: 1194.79ms | tok/sec: 877,622 | bf16_mfu: 53.91 | total time: 48.49m | eta: 62.4m step 02440/05568 (43.82%) | loss: 2.613278 | lrm: 0.87 | dt: 1189.43ms | tok/sec: 881,581 | bf16_mfu: 54.15 | total time: 48.51m | eta: 62.4m step 02441/05568 (43.84%) | loss: 2.613912 | lrm: 0.87 | dt: 1189.91ms | tok/sec: 881,226 | bf16_mfu: 54.13 | total time: 48.53m | eta: 62.4m step 02442/05568 (43.86%) | loss: 2.612921 | lrm: 0.87 | dt: 1192.66ms | tok/sec: 879,191 | bf16_mfu: 54.01 | total time: 48.55m | eta: 62.4m step 02443/05568 (43.88%) | loss: 2.614903 | lrm: 0.87 | dt: 1191.05ms | tok/sec: 880,379 | bf16_mfu: 54.08 | total time: 48.57m | eta: 62.4m step 02444/05568 (43.89%) | loss: 2.601782 | lrm: 0.87 | dt: 1190.15ms | tok/sec: 881,043 | bf16_mfu: 54.12 | total time: 48.59m | eta: 62.3m step 02445/05568 (43.91%) | loss: 2.589780 | lrm: 0.87 | dt: 1192.16ms | tok/sec: 879,563 | bf16_mfu: 54.03 | total time: 48.61m | eta: 62.3m step 02446/05568 (43.93%) | loss: 2.577083 | lrm: 0.87 | dt: 1192.29ms | tok/sec: 879,465 | bf16_mfu: 54.02 | total time: 48.63m | eta: 62.3m step 02447/05568 (43.95%) | loss: 2.582171 | lrm: 0.87 | dt: 1189.63ms | tok/sec: 881,432 | bf16_mfu: 54.14 | total time: 48.65m | eta: 62.3m step 02448/05568 (43.97%) | loss: 2.589279 | lrm: 0.87 | dt: 1192.95ms | tok/sec: 878,976 | bf16_mfu: 53.99 | total time: 48.67m | eta: 62.3m step 02449/05568 (43.98%) | loss: 2.590885 | lrm: 0.87 | dt: 1188.79ms | tok/sec: 882,050 | bf16_mfu: 54.18 | total time: 48.69m | eta: 62.2m step 02450/05568 (44.00%) | loss: 2.591251 | lrm: 0.87 | dt: 1192.07ms | tok/sec: 879,625 | bf16_mfu: 54.03 | total time: 48.71m | eta: 62.2m step 02451/05568 (44.02%) | loss: 2.596435 | lrm: 0.87 | dt: 1188.07ms | tok/sec: 882,586 | bf16_mfu: 54.22 | total time: 48.73m | eta: 62.2m step 02452/05568 (44.04%) | loss: 2.603061 | lrm: 0.87 | dt: 1192.29ms | tok/sec: 879,464 | bf16_mfu: 54.02 | total time: 48.75m | eta: 62.2m step 02453/05568 (44.06%) | loss: 2.602472 | lrm: 0.87 | dt: 1190.10ms | tok/sec: 881,085 | bf16_mfu: 54.12 | total time: 48.77m | eta: 62.2m step 02454/05568 (44.07%) | loss: 2.603160 | lrm: 0.87 | dt: 1192.51ms | tok/sec: 879,297 | bf16_mfu: 54.01 | total time: 48.79m | eta: 62.1m step 02455/05568 (44.09%) | loss: 2.602019 | lrm: 0.87 | dt: 1194.93ms | tok/sec: 877,518 | bf16_mfu: 53.90 | total time: 48.81m | eta: 62.1m step 02456/05568 (44.11%) | loss: 2.604674 | lrm: 0.87 | dt: 1191.94ms | tok/sec: 879,720 | bf16_mfu: 54.04 | total time: 48.83m | eta: 62.1m step 02457/05568 (44.13%) | loss: 2.611739 | lrm: 0.87 | dt: 1190.52ms | tok/sec: 880,767 | bf16_mfu: 54.10 | total time: 48.85m | eta: 62.1m step 02458/05568 (44.15%) | loss: 2.616917 | lrm: 0.87 | dt: 1193.27ms | tok/sec: 878,744 | bf16_mfu: 53.98 | total time: 48.87m | eta: 62.1m step 02459/05568 (44.16%) | loss: 2.623228 | lrm: 0.87 | dt: 1192.84ms | tok/sec: 879,060 | bf16_mfu: 54.00 | total time: 48.89m | eta: 62.0m step 02460/05568 (44.18%) | loss: 2.631794 | lrm: 0.87 | dt: 1189.49ms | tok/sec: 881,534 | bf16_mfu: 54.15 | total time: 48.91m | eta: 62.0m step 02461/05568 (44.20%) | loss: 2.632721 | lrm: 0.87 | dt: 1195.36ms | tok/sec: 877,204 | bf16_mfu: 53.88 | total time: 48.93m | eta: 62.0m step 02462/05568 (44.22%) | loss: 2.641340 | lrm: 0.87 | dt: 1190.01ms | tok/sec: 881,151 | bf16_mfu: 54.13 | total time: 48.95m | eta: 62.0m step 02463/05568 (44.23%) | loss: 2.632609 | lrm: 0.87 | dt: 1193.11ms | tok/sec: 878,862 | bf16_mfu: 53.99 | total time: 48.97m | eta: 62.0m step 02464/05568 (44.25%) | loss: 2.629161 | lrm: 0.86 | dt: 1191.82ms | tok/sec: 879,811 | bf16_mfu: 54.04 | total time: 48.99m | eta: 61.9m step 02465/05568 (44.27%) | loss: 2.621442 | lrm: 0.86 | dt: 1191.81ms | tok/sec: 879,817 | bf16_mfu: 54.05 | total time: 49.01m | eta: 61.9m step 02466/05568 (44.29%) | loss: 2.620915 | lrm: 0.86 | dt: 1189.77ms | tok/sec: 881,329 | bf16_mfu: 54.14 | total time: 49.03m | eta: 61.9m step 02467/05568 (44.31%) | loss: 2.618507 | lrm: 0.86 | dt: 1192.17ms | tok/sec: 879,550 | bf16_mfu: 54.03 | total time: 49.05m | eta: 61.9m step 02468/05568 (44.32%) | loss: 2.612779 | lrm: 0.86 | dt: 1188.57ms | tok/sec: 882,216 | bf16_mfu: 54.19 | total time: 49.07m | eta: 61.9m step 02469/05568 (44.34%) | loss: 2.606599 | lrm: 0.86 | dt: 1187.85ms | tok/sec: 882,752 | bf16_mfu: 54.23 | total time: 49.09m | eta: 61.8m step 02470/05568 (44.36%) | loss: 2.610741 | lrm: 0.86 | dt: 1194.30ms | tok/sec: 877,985 | bf16_mfu: 53.93 | total time: 49.11m | eta: 61.8m step 02471/05568 (44.38%) | loss: 2.622541 | lrm: 0.86 | dt: 1190.09ms | tok/sec: 881,089 | bf16_mfu: 54.12 | total time: 49.13m | eta: 61.8m step 02472/05568 (44.40%) | loss: 2.607037 | lrm: 0.86 | dt: 1190.39ms | tok/sec: 880,866 | bf16_mfu: 54.11 | total time: 49.15m | eta: 61.8m step 02473/05568 (44.41%) | loss: 2.601401 | lrm: 0.86 | dt: 1189.70ms | tok/sec: 881,378 | bf16_mfu: 54.14 | total time: 49.17m | eta: 61.8m step 02474/05568 (44.43%) | loss: 2.604589 | lrm: 0.86 | dt: 1189.89ms | tok/sec: 881,235 | bf16_mfu: 54.13 | total time: 49.19m | eta: 61.7m step 02475/05568 (44.45%) | loss: 2.611884 | lrm: 0.86 | dt: 1191.69ms | tok/sec: 879,909 | bf16_mfu: 54.05 | total time: 49.21m | eta: 61.7m step 02476/05568 (44.47%) | loss: 2.604670 | lrm: 0.86 | dt: 1190.58ms | tok/sec: 880,729 | bf16_mfu: 54.10 | total time: 49.23m | eta: 61.7m step 02477/05568 (44.49%) | loss: 2.604568 | lrm: 0.86 | dt: 1193.85ms | tok/sec: 878,312 | bf16_mfu: 53.95 | total time: 49.25m | eta: 61.7m step 02478/05568 (44.50%) | loss: 2.613455 | lrm: 0.86 | dt: 1190.36ms | tok/sec: 880,887 | bf16_mfu: 54.11 | total time: 49.27m | eta: 61.7m step 02479/05568 (44.52%) | loss: 2.618851 | lrm: 0.86 | dt: 1191.45ms | tok/sec: 880,085 | bf16_mfu: 54.06 | total time: 49.29m | eta: 61.6m step 02480/05568 (44.54%) | loss: 2.622377 | lrm: 0.86 | dt: 1192.42ms | tok/sec: 879,364 | bf16_mfu: 54.02 | total time: 49.31m | eta: 61.6m step 02481/05568 (44.56%) | loss: 2.616547 | lrm: 0.86 | dt: 1190.12ms | tok/sec: 881,064 | bf16_mfu: 54.12 | total time: 49.33m | eta: 61.6m step 02482/05568 (44.58%) | loss: 2.625800 | lrm: 0.86 | dt: 1192.04ms | tok/sec: 879,648 | bf16_mfu: 54.03 | total time: 49.35m | eta: 61.6m step 02483/05568 (44.59%) | loss: 2.626747 | lrm: 0.86 | dt: 1194.66ms | tok/sec: 877,716 | bf16_mfu: 53.92 | total time: 49.37m | eta: 61.6m step 02484/05568 (44.61%) | loss: 2.627158 | lrm: 0.86 | dt: 1192.75ms | tok/sec: 879,127 | bf16_mfu: 54.00 | total time: 49.39m | eta: 61.5m step 02485/05568 (44.63%) | loss: 2.631577 | lrm: 0.86 | dt: 1194.15ms | tok/sec: 878,090 | bf16_mfu: 53.94 | total time: 49.41m | eta: 61.5m step 02486/05568 (44.65%) | loss: 2.638776 | lrm: 0.86 | dt: 1193.77ms | tok/sec: 878,372 | bf16_mfu: 53.96 | total time: 49.43m | eta: 61.5m step 02487/05568 (44.67%) | loss: 2.629972 | lrm: 0.86 | dt: 1189.82ms | tok/sec: 881,291 | bf16_mfu: 54.14 | total time: 49.45m | eta: 61.5m step 02488/05568 (44.68%) | loss: 2.623618 | lrm: 0.86 | dt: 1195.94ms | tok/sec: 876,783 | bf16_mfu: 53.86 | total time: 49.46m | eta: 61.5m step 02489/05568 (44.70%) | loss: 2.617182 | lrm: 0.86 | dt: 1188.63ms | tok/sec: 882,173 | bf16_mfu: 54.19 | total time: 49.48m | eta: 61.4m step 02490/05568 (44.72%) | loss: 2.621986 | lrm: 0.86 | dt: 1192.08ms | tok/sec: 879,617 | bf16_mfu: 54.03 | total time: 49.50m | eta: 61.4m step 02491/05568 (44.74%) | loss: 2.623534 | lrm: 0.86 | dt: 1188.87ms | tok/sec: 881,993 | bf16_mfu: 54.18 | total time: 49.52m | eta: 61.4m step 02492/05568 (44.76%) | loss: 2.627483 | lrm: 0.86 | dt: 1189.04ms | tok/sec: 881,868 | bf16_mfu: 54.17 | total time: 49.54m | eta: 61.4m step 02493/05568 (44.77%) | loss: 2.631258 | lrm: 0.86 | dt: 1195.62ms | tok/sec: 877,012 | bf16_mfu: 53.87 | total time: 49.56m | eta: 61.4m step 02494/05568 (44.79%) | loss: 2.618222 | lrm: 0.86 | dt: 1194.52ms | tok/sec: 877,819 | bf16_mfu: 53.92 | total time: 49.58m | eta: 61.3m step 02495/05568 (44.81%) | loss: 2.617375 | lrm: 0.86 | dt: 1195.66ms | tok/sec: 876,982 | bf16_mfu: 53.87 | total time: 49.60m | eta: 61.3m step 02496/05568 (44.83%) | loss: 2.616441 | lrm: 0.86 | dt: 1192.47ms | tok/sec: 879,332 | bf16_mfu: 54.02 | total time: 49.62m | eta: 61.3m step 02497/05568 (44.85%) | loss: 2.609997 | lrm: 0.86 | dt: 1195.05ms | tok/sec: 877,430 | bf16_mfu: 53.90 | total time: 49.64m | eta: 61.3m step 02498/05568 (44.86%) | loss: 2.606499 | lrm: 0.86 | dt: 1192.89ms | tok/sec: 879,022 | bf16_mfu: 54.00 | total time: 49.66m | eta: 61.3m step 02499/05568 (44.88%) | loss: 2.611066 | lrm: 0.86 | dt: 1193.95ms | tok/sec: 878,241 | bf16_mfu: 53.95 | total time: 49.68m | eta: 61.2m step:2500/5568 val_bpb:0.789216 val_time:3.90s step 02500/05568 (44.90%) | loss: 2.622232 | lrm: 0.86 | dt: 1191.10ms | tok/sec: 880,340 | bf16_mfu: 54.08 | total time: 49.70m | eta: 61.2m step 02501/05568 (44.92%) | loss: 2.629286 | lrm: 0.86 | dt: 1190.56ms | tok/sec: 880,739 | bf16_mfu: 54.10 | total time: 49.72m | eta: 61.2m step 02502/05568 (44.94%) | loss: 2.627412 | lrm: 0.85 | dt: 1193.60ms | tok/sec: 878,499 | bf16_mfu: 53.96 | total time: 49.74m | eta: 61.2m step 02503/05568 (44.95%) | loss: 2.618120 | lrm: 0.85 | dt: 1191.45ms | tok/sec: 880,082 | bf16_mfu: 54.06 | total time: 49.76m | eta: 61.2m step 02504/05568 (44.97%) | loss: 2.613961 | lrm: 0.85 | dt: 1193.44ms | tok/sec: 878,617 | bf16_mfu: 53.97 | total time: 49.78m | eta: 61.1m step 02505/05568 (44.99%) | loss: 2.608845 | lrm: 0.85 | dt: 1190.31ms | tok/sec: 880,929 | bf16_mfu: 54.11 | total time: 49.80m | eta: 61.1m step 02506/05568 (45.01%) | loss: 2.610317 | lrm: 0.85 | dt: 1188.94ms | tok/sec: 881,941 | bf16_mfu: 54.18 | total time: 49.82m | eta: 61.1m step 02507/05568 (45.03%) | loss: 2.608627 | lrm: 0.85 | dt: 1192.35ms | tok/sec: 879,421 | bf16_mfu: 54.02 | total time: 49.84m | eta: 61.1m step 02508/05568 (45.04%) | loss: 2.608870 | lrm: 0.85 | dt: 1195.13ms | tok/sec: 877,375 | bf16_mfu: 53.90 | total time: 49.86m | eta: 61.1m step 02509/05568 (45.06%) | loss: 2.609891 | lrm: 0.85 | dt: 1190.04ms | tok/sec: 881,129 | bf16_mfu: 54.13 | total time: 49.88m | eta: 61.0m step 02510/05568 (45.08%) | loss: 2.607731 | lrm: 0.85 | dt: 1191.76ms | tok/sec: 879,851 | bf16_mfu: 54.05 | total time: 49.90m | eta: 61.0m step 02511/05568 (45.10%) | loss: 2.596035 | lrm: 0.85 | dt: 1193.90ms | tok/sec: 878,279 | bf16_mfu: 53.95 | total time: 49.92m | eta: 61.0m step 02512/05568 (45.11%) | loss: 2.609404 | lrm: 0.85 | dt: 1192.58ms | tok/sec: 879,253 | bf16_mfu: 54.01 | total time: 49.94m | eta: 61.0m step 02513/05568 (45.13%) | loss: 2.603270 | lrm: 0.85 | dt: 1191.49ms | tok/sec: 880,051 | bf16_mfu: 54.06 | total time: 49.96m | eta: 61.0m step 02514/05568 (45.15%) | loss: 2.598313 | lrm: 0.85 | dt: 1192.79ms | tok/sec: 879,092 | bf16_mfu: 54.00 | total time: 49.98m | eta: 60.9m step 02515/05568 (45.17%) | loss: 2.600293 | lrm: 0.85 | dt: 1186.68ms | tok/sec: 883,620 | bf16_mfu: 54.28 | total time: 50.00m | eta: 60.9m step 02516/05568 (45.19%) | loss: 2.594983 | lrm: 0.85 | dt: 1188.05ms | tok/sec: 882,599 | bf16_mfu: 54.22 | total time: 50.02m | eta: 60.9m step 02517/05568 (45.20%) | loss: 2.580856 | lrm: 0.85 | dt: 1191.33ms | tok/sec: 880,170 | bf16_mfu: 54.07 | total time: 50.04m | eta: 60.9m step 02518/05568 (45.22%) | loss: 2.590960 | lrm: 0.85 | dt: 1188.94ms | tok/sec: 881,939 | bf16_mfu: 54.18 | total time: 50.06m | eta: 60.9m step 02519/05568 (45.24%) | loss: 2.586711 | lrm: 0.85 | dt: 1192.55ms | tok/sec: 879,271 | bf16_mfu: 54.01 | total time: 50.08m | eta: 60.8m step 02520/05568 (45.26%) | loss: 2.587995 | lrm: 0.85 | dt: 1193.04ms | tok/sec: 878,911 | bf16_mfu: 53.99 | total time: 50.10m | eta: 60.8m step 02521/05568 (45.28%) | loss: 2.593006 | lrm: 0.85 | dt: 1194.02ms | tok/sec: 878,187 | bf16_mfu: 53.95 | total time: 50.12m | eta: 60.8m step 02522/05568 (45.29%) | loss: 2.596396 | lrm: 0.85 | dt: 1194.20ms | tok/sec: 878,058 | bf16_mfu: 53.94 | total time: 50.14m | eta: 60.8m step 02523/05568 (45.31%) | loss: 2.586397 | lrm: 0.85 | dt: 1194.69ms | tok/sec: 877,700 | bf16_mfu: 53.92 | total time: 50.16m | eta: 60.8m step 02524/05568 (45.33%) | loss: 2.577791 | lrm: 0.85 | dt: 1196.87ms | tok/sec: 876,097 | bf16_mfu: 53.82 | total time: 50.18m | eta: 60.7m step 02525/05568 (45.35%) | loss: 2.583952 | lrm: 0.85 | dt: 1193.97ms | tok/sec: 878,227 | bf16_mfu: 53.95 | total time: 50.20m | eta: 60.7m step 02526/05568 (45.37%) | loss: 2.591603 | lrm: 0.85 | dt: 1190.15ms | tok/sec: 881,042 | bf16_mfu: 54.12 | total time: 50.22m | eta: 60.7m step 02527/05568 (45.38%) | loss: 2.595406 | lrm: 0.85 | dt: 1192.71ms | tok/sec: 879,155 | bf16_mfu: 54.00 | total time: 50.24m | eta: 60.7m step 02528/05568 (45.40%) | loss: 2.598100 | lrm: 0.85 | dt: 1193.82ms | tok/sec: 878,334 | bf16_mfu: 53.95 | total time: 50.26m | eta: 60.7m step 02529/05568 (45.42%) | loss: 2.590943 | lrm: 0.85 | dt: 1189.44ms | tok/sec: 881,568 | bf16_mfu: 54.15 | total time: 50.28m | eta: 60.6m step 02530/05568 (45.44%) | loss: 2.592240 | lrm: 0.85 | dt: 1195.13ms | tok/sec: 877,370 | bf16_mfu: 53.89 | total time: 50.30m | eta: 60.6m step 02531/05568 (45.46%) | loss: 2.593817 | lrm: 0.85 | dt: 1189.10ms | tok/sec: 881,822 | bf16_mfu: 54.17 | total time: 50.32m | eta: 60.6m step 02532/05568 (45.47%) | loss: 2.593156 | lrm: 0.85 | dt: 1194.12ms | tok/sec: 878,113 | bf16_mfu: 53.94 | total time: 50.34m | eta: 60.6m step 02533/05568 (45.49%) | loss: 2.592080 | lrm: 0.85 | dt: 1193.03ms | tok/sec: 878,919 | bf16_mfu: 53.99 | total time: 50.36m | eta: 60.6m step 02534/05568 (45.51%) | loss: 2.590437 | lrm: 0.85 | dt: 1192.41ms | tok/sec: 879,374 | bf16_mfu: 54.02 | total time: 50.38m | eta: 60.5m step 02535/05568 (45.53%) | loss: 2.581155 | lrm: 0.85 | dt: 1189.50ms | tok/sec: 881,528 | bf16_mfu: 54.15 | total time: 50.40m | eta: 60.5m step 02536/05568 (45.55%) | loss: 2.571558 | lrm: 0.85 | dt: 1190.33ms | tok/sec: 880,914 | bf16_mfu: 54.11 | total time: 50.42m | eta: 60.5m step 02537/05568 (45.56%) | loss: 2.566596 | lrm: 0.85 | dt: 1197.29ms | tok/sec: 875,794 | bf16_mfu: 53.80 | total time: 50.44m | eta: 60.5m step 02538/05568 (45.58%) | loss: 2.575847 | lrm: 0.85 | dt: 1192.59ms | tok/sec: 879,243 | bf16_mfu: 54.01 | total time: 50.46m | eta: 60.5m step 02539/05568 (45.60%) | loss: 2.576463 | lrm: 0.85 | dt: 1187.84ms | tok/sec: 882,758 | bf16_mfu: 54.23 | total time: 50.48m | eta: 60.4m step 02540/05568 (45.62%) | loss: 2.572214 | lrm: 0.84 | dt: 1192.94ms | tok/sec: 878,987 | bf16_mfu: 53.99 | total time: 50.50m | eta: 60.4m step 02541/05568 (45.64%) | loss: 2.575350 | lrm: 0.84 | dt: 1189.34ms | tok/sec: 881,647 | bf16_mfu: 54.16 | total time: 50.52m | eta: 60.4m step 02542/05568 (45.65%) | loss: 2.579883 | lrm: 0.84 | dt: 1195.73ms | tok/sec: 876,930 | bf16_mfu: 53.87 | total time: 50.54m | eta: 60.4m step 02543/05568 (45.67%) | loss: 2.575527 | lrm: 0.84 | dt: 1192.26ms | tok/sec: 879,485 | bf16_mfu: 54.02 | total time: 50.56m | eta: 60.4m step 02544/05568 (45.69%) | loss: 2.571955 | lrm: 0.84 | dt: 1186.58ms | tok/sec: 883,699 | bf16_mfu: 54.28 | total time: 50.58m | eta: 60.3m step 02545/05568 (45.71%) | loss: 2.568131 | lrm: 0.84 | dt: 1192.61ms | tok/sec: 879,227 | bf16_mfu: 54.01 | total time: 50.60m | eta: 60.3m step 02546/05568 (45.73%) | loss: 2.566272 | lrm: 0.84 | dt: 1190.81ms | tok/sec: 880,556 | bf16_mfu: 54.09 | total time: 50.62m | eta: 60.3m step 02547/05568 (45.74%) | loss: 2.571686 | lrm: 0.84 | dt: 1194.92ms | tok/sec: 877,531 | bf16_mfu: 53.90 | total time: 50.64m | eta: 60.3m step 02548/05568 (45.76%) | loss: 2.561086 | lrm: 0.84 | dt: 1195.73ms | tok/sec: 876,933 | bf16_mfu: 53.87 | total time: 50.66m | eta: 60.3m step 02549/05568 (45.78%) | loss: 2.571876 | lrm: 0.84 | dt: 1192.88ms | tok/sec: 879,026 | bf16_mfu: 54.00 | total time: 50.68m | eta: 60.2m step 02550/05568 (45.80%) | loss: 2.580051 | lrm: 0.84 | dt: 1191.67ms | tok/sec: 879,923 | bf16_mfu: 54.05 | total time: 50.70m | eta: 60.2m step 02551/05568 (45.82%) | loss: 2.575533 | lrm: 0.84 | dt: 1193.31ms | tok/sec: 878,709 | bf16_mfu: 53.98 | total time: 50.72m | eta: 60.2m step 02552/05568 (45.83%) | loss: 2.581244 | lrm: 0.84 | dt: 1192.25ms | tok/sec: 879,495 | bf16_mfu: 54.03 | total time: 50.74m | eta: 60.2m step 02553/05568 (45.85%) | loss: 2.592168 | lrm: 0.84 | dt: 1191.88ms | tok/sec: 879,767 | bf16_mfu: 54.04 | total time: 50.76m | eta: 60.2m step 02554/05568 (45.87%) | loss: 2.578392 | lrm: 0.84 | dt: 1197.51ms | tok/sec: 875,632 | bf16_mfu: 53.79 | total time: 50.78m | eta: 60.1m step 02555/05568 (45.89%) | loss: 2.585854 | lrm: 0.84 | dt: 1201.77ms | tok/sec: 872,524 | bf16_mfu: 53.60 | total time: 50.80m | eta: 60.1m step 02556/05568 (45.91%) | loss: 2.585806 | lrm: 0.84 | dt: 1189.69ms | tok/sec: 881,388 | bf16_mfu: 54.14 | total time: 50.82m | eta: 60.1m step 02557/05568 (45.92%) | loss: 2.590669 | lrm: 0.84 | dt: 1190.88ms | tok/sec: 880,502 | bf16_mfu: 54.09 | total time: 50.84m | eta: 60.1m step 02558/05568 (45.94%) | loss: 2.591380 | lrm: 0.84 | dt: 1192.52ms | tok/sec: 879,296 | bf16_mfu: 54.01 | total time: 50.86m | eta: 60.1m step 02559/05568 (45.96%) | loss: 2.582442 | lrm: 0.84 | dt: 1189.78ms | tok/sec: 881,317 | bf16_mfu: 54.14 | total time: 50.88m | eta: 60.0m step 02560/05568 (45.98%) | loss: 2.575142 | lrm: 0.84 | dt: 1190.12ms | tok/sec: 881,064 | bf16_mfu: 54.12 | total time: 50.90m | eta: 60.0m step 02561/05568 (45.99%) | loss: 2.569189 | lrm: 0.84 | dt: 1193.16ms | tok/sec: 878,820 | bf16_mfu: 53.98 | total time: 50.92m | eta: 60.0m step 02562/05568 (46.01%) | loss: 2.577638 | lrm: 0.84 | dt: 1190.63ms | tok/sec: 880,690 | bf16_mfu: 54.10 | total time: 50.94m | eta: 60.0m step 02563/05568 (46.03%) | loss: 2.580195 | lrm: 0.84 | dt: 1192.17ms | tok/sec: 879,552 | bf16_mfu: 54.03 | total time: 50.96m | eta: 60.0m step 02564/05568 (46.05%) | loss: 2.581383 | lrm: 0.84 | dt: 1196.25ms | tok/sec: 876,552 | bf16_mfu: 53.84 | total time: 50.98m | eta: 59.9m step 02565/05568 (46.07%) | loss: 2.582088 | lrm: 0.84 | dt: 1184.22ms | tok/sec: 885,457 | bf16_mfu: 54.39 | total time: 50.99m | eta: 59.9m step 02566/05568 (46.08%) | loss: 2.579357 | lrm: 0.84 | dt: 1187.55ms | tok/sec: 882,973 | bf16_mfu: 54.24 | total time: 51.01m | eta: 59.9m step 02567/05568 (46.10%) | loss: 2.578084 | lrm: 0.84 | dt: 1192.09ms | tok/sec: 879,608 | bf16_mfu: 54.03 | total time: 51.03m | eta: 59.9m step 02568/05568 (46.12%) | loss: 2.577453 | lrm: 0.84 | dt: 1190.12ms | tok/sec: 881,068 | bf16_mfu: 54.12 | total time: 51.05m | eta: 59.9m step 02569/05568 (46.14%) | loss: 2.577491 | lrm: 0.84 | dt: 1188.93ms | tok/sec: 881,946 | bf16_mfu: 54.18 | total time: 51.07m | eta: 59.8m step 02570/05568 (46.16%) | loss: 2.573599 | lrm: 0.84 | dt: 1189.81ms | tok/sec: 881,293 | bf16_mfu: 54.14 | total time: 51.09m | eta: 59.8m step 02571/05568 (46.17%) | loss: 2.575244 | lrm: 0.84 | dt: 1187.66ms | tok/sec: 882,889 | bf16_mfu: 54.23 | total time: 51.11m | eta: 59.8m step 02572/05568 (46.19%) | loss: 2.565119 | lrm: 0.84 | dt: 1190.37ms | tok/sec: 880,882 | bf16_mfu: 54.11 | total time: 51.13m | eta: 59.8m step 02573/05568 (46.21%) | loss: 2.565486 | lrm: 0.84 | dt: 1190.02ms | tok/sec: 881,140 | bf16_mfu: 54.13 | total time: 51.15m | eta: 59.8m step 02574/05568 (46.23%) | loss: 2.559397 | lrm: 0.84 | dt: 1188.45ms | tok/sec: 882,308 | bf16_mfu: 54.20 | total time: 51.17m | eta: 59.7m step 02575/05568 (46.25%) | loss: 2.557060 | lrm: 0.84 | dt: 1196.42ms | tok/sec: 876,424 | bf16_mfu: 53.84 | total time: 51.19m | eta: 59.7m step 02576/05568 (46.26%) | loss: 2.572736 | lrm: 0.84 | dt: 1189.72ms | tok/sec: 881,361 | bf16_mfu: 54.14 | total time: 51.21m | eta: 59.7m step 02577/05568 (46.28%) | loss: 2.577202 | lrm: 0.84 | dt: 1194.04ms | tok/sec: 878,173 | bf16_mfu: 53.94 | total time: 51.23m | eta: 59.7m step 02578/05568 (46.30%) | loss: 2.571223 | lrm: 0.83 | dt: 1194.59ms | tok/sec: 877,767 | bf16_mfu: 53.92 | total time: 51.25m | eta: 59.7m step 02579/05568 (46.32%) | loss: 2.574146 | lrm: 0.83 | dt: 1188.02ms | tok/sec: 882,625 | bf16_mfu: 54.22 | total time: 51.27m | eta: 59.6m step 02580/05568 (46.34%) | loss: 2.570255 | lrm: 0.83 | dt: 1195.50ms | tok/sec: 877,105 | bf16_mfu: 53.88 | total time: 51.29m | eta: 59.6m step 02581/05568 (46.35%) | loss: 2.575457 | lrm: 0.83 | dt: 1192.18ms | tok/sec: 879,541 | bf16_mfu: 54.03 | total time: 51.31m | eta: 59.6m step 02582/05568 (46.37%) | loss: 2.576888 | lrm: 0.83 | dt: 1190.24ms | tok/sec: 880,977 | bf16_mfu: 54.12 | total time: 51.33m | eta: 59.6m step 02583/05568 (46.39%) | loss: 2.578012 | lrm: 0.83 | dt: 1192.67ms | tok/sec: 879,186 | bf16_mfu: 54.01 | total time: 51.35m | eta: 59.6m step 02584/05568 (46.41%) | loss: 2.571819 | lrm: 0.83 | dt: 1191.29ms | tok/sec: 880,203 | bf16_mfu: 54.07 | total time: 51.37m | eta: 59.5m step 02585/05568 (46.43%) | loss: 2.573770 | lrm: 0.83 | dt: 1189.02ms | tok/sec: 881,886 | bf16_mfu: 54.17 | total time: 51.39m | eta: 59.5m step 02586/05568 (46.44%) | loss: 2.573748 | lrm: 0.83 | dt: 1191.47ms | tok/sec: 880,071 | bf16_mfu: 54.06 | total time: 51.41m | eta: 59.5m step 02587/05568 (46.46%) | loss: 2.587258 | lrm: 0.83 | dt: 1194.67ms | tok/sec: 877,711 | bf16_mfu: 53.92 | total time: 51.43m | eta: 59.5m step 02588/05568 (46.48%) | loss: 2.584890 | lrm: 0.83 | dt: 1194.23ms | tok/sec: 878,035 | bf16_mfu: 53.94 | total time: 51.45m | eta: 59.5m step 02589/05568 (46.50%) | loss: 2.589807 | lrm: 0.83 | dt: 1190.31ms | tok/sec: 880,924 | bf16_mfu: 54.11 | total time: 51.47m | eta: 59.4m step 02590/05568 (46.52%) | loss: 2.567432 | lrm: 0.83 | dt: 1195.04ms | tok/sec: 877,440 | bf16_mfu: 53.90 | total time: 51.49m | eta: 59.4m step 02591/05568 (46.53%) | loss: 2.572799 | lrm: 0.83 | dt: 1193.18ms | tok/sec: 878,808 | bf16_mfu: 53.98 | total time: 51.51m | eta: 59.4m step 02592/05568 (46.55%) | loss: 2.587114 | lrm: 0.83 | dt: 1193.00ms | tok/sec: 878,937 | bf16_mfu: 53.99 | total time: 51.53m | eta: 59.4m step 02593/05568 (46.57%) | loss: 2.592395 | lrm: 0.83 | dt: 1193.60ms | tok/sec: 878,498 | bf16_mfu: 53.96 | total time: 51.55m | eta: 59.4m step 02594/05568 (46.59%) | loss: 2.596732 | lrm: 0.83 | dt: 1197.19ms | tok/sec: 875,865 | bf16_mfu: 53.80 | total time: 51.57m | eta: 59.3m step 02595/05568 (46.61%) | loss: 2.601518 | lrm: 0.83 | dt: 1194.55ms | tok/sec: 877,801 | bf16_mfu: 53.92 | total time: 51.59m | eta: 59.3m step 02596/05568 (46.62%) | loss: 2.591572 | lrm: 0.83 | dt: 1190.49ms | tok/sec: 880,796 | bf16_mfu: 54.11 | total time: 51.61m | eta: 59.3m step 02597/05568 (46.64%) | loss: 2.586811 | lrm: 0.83 | dt: 1193.42ms | tok/sec: 878,633 | bf16_mfu: 53.97 | total time: 51.63m | eta: 59.3m step 02598/05568 (46.66%) | loss: 2.586891 | lrm: 0.83 | dt: 1190.22ms | tok/sec: 880,995 | bf16_mfu: 54.12 | total time: 51.65m | eta: 59.3m step 02599/05568 (46.68%) | loss: 2.580521 | lrm: 0.83 | dt: 1192.53ms | tok/sec: 879,287 | bf16_mfu: 54.01 | total time: 51.67m | eta: 59.2m step 02600/05568 (46.70%) | loss: 2.573247 | lrm: 0.83 | dt: 1191.94ms | tok/sec: 879,722 | bf16_mfu: 54.04 | total time: 51.69m | eta: 59.2m step 02601/05568 (46.71%) | loss: 2.571024 | lrm: 0.83 | dt: 1193.64ms | tok/sec: 878,472 | bf16_mfu: 53.96 | total time: 51.71m | eta: 59.2m step 02602/05568 (46.73%) | loss: 2.578279 | lrm: 0.83 | dt: 1189.91ms | tok/sec: 881,220 | bf16_mfu: 54.13 | total time: 51.73m | eta: 59.2m step 02603/05568 (46.75%) | loss: 2.587683 | lrm: 0.83 | dt: 1195.66ms | tok/sec: 876,983 | bf16_mfu: 53.87 | total time: 51.75m | eta: 59.2m step 02604/05568 (46.77%) | loss: 2.583126 | lrm: 0.83 | dt: 1186.17ms | tok/sec: 884,000 | bf16_mfu: 54.30 | total time: 51.77m | eta: 59.1m step 02605/05568 (46.79%) | loss: 2.585878 | lrm: 0.83 | dt: 1193.26ms | tok/sec: 878,748 | bf16_mfu: 53.98 | total time: 51.79m | eta: 59.1m step 02606/05568 (46.80%) | loss: 2.582791 | lrm: 0.83 | dt: 1191.89ms | tok/sec: 879,760 | bf16_mfu: 54.04 | total time: 51.81m | eta: 59.1m step 02607/05568 (46.82%) | loss: 2.593158 | lrm: 0.83 | dt: 1192.05ms | tok/sec: 879,641 | bf16_mfu: 54.03 | total time: 51.83m | eta: 59.1m step 02608/05568 (46.84%) | loss: 2.587307 | lrm: 0.83 | dt: 1191.11ms | tok/sec: 880,336 | bf16_mfu: 54.08 | total time: 51.85m | eta: 59.1m step 02609/05568 (46.86%) | loss: 2.584237 | lrm: 0.83 | dt: 1189.74ms | tok/sec: 881,351 | bf16_mfu: 54.14 | total time: 51.87m | eta: 59.0m step 02610/05568 (46.88%) | loss: 2.579774 | lrm: 0.83 | dt: 1192.40ms | tok/sec: 879,385 | bf16_mfu: 54.02 | total time: 51.89m | eta: 59.0m step 02611/05568 (46.89%) | loss: 2.586214 | lrm: 0.83 | dt: 1191.46ms | tok/sec: 880,077 | bf16_mfu: 54.06 | total time: 51.91m | eta: 59.0m step 02612/05568 (46.91%) | loss: 2.586919 | lrm: 0.83 | dt: 1195.41ms | tok/sec: 877,169 | bf16_mfu: 53.88 | total time: 51.93m | eta: 59.0m step 02613/05568 (46.93%) | loss: 2.584773 | lrm: 0.83 | dt: 1190.89ms | tok/sec: 880,500 | bf16_mfu: 54.09 | total time: 51.95m | eta: 59.0m step 02614/05568 (46.95%) | loss: 2.581965 | lrm: 0.83 | dt: 1195.41ms | tok/sec: 877,170 | bf16_mfu: 53.88 | total time: 51.97m | eta: 58.9m step 02615/05568 (46.96%) | loss: 2.585895 | lrm: 0.83 | dt: 1194.19ms | tok/sec: 878,061 | bf16_mfu: 53.94 | total time: 51.99m | eta: 58.9m step 02616/05568 (46.98%) | loss: 2.593929 | lrm: 0.82 | dt: 1195.69ms | tok/sec: 876,964 | bf16_mfu: 53.87 | total time: 52.01m | eta: 58.9m step 02617/05568 (47.00%) | loss: 2.599537 | lrm: 0.82 | dt: 1192.41ms | tok/sec: 879,372 | bf16_mfu: 54.02 | total time: 52.03m | eta: 58.9m step 02618/05568 (47.02%) | loss: 2.596198 | lrm: 0.82 | dt: 1192.30ms | tok/sec: 879,453 | bf16_mfu: 54.02 | total time: 52.05m | eta: 58.9m step 02619/05568 (47.04%) | loss: 2.577320 | lrm: 0.82 | dt: 1192.54ms | tok/sec: 879,282 | bf16_mfu: 54.01 | total time: 52.07m | eta: 58.8m step 02620/05568 (47.05%) | loss: 2.571824 | lrm: 0.82 | dt: 1190.05ms | tok/sec: 881,115 | bf16_mfu: 54.12 | total time: 52.09m | eta: 58.8m step 02621/05568 (47.07%) | loss: 2.566413 | lrm: 0.82 | dt: 1191.98ms | tok/sec: 879,694 | bf16_mfu: 54.04 | total time: 52.11m | eta: 58.8m step 02622/05568 (47.09%) | loss: 2.571874 | lrm: 0.82 | dt: 1190.40ms | tok/sec: 880,861 | bf16_mfu: 54.11 | total time: 52.13m | eta: 58.8m step 02623/05568 (47.11%) | loss: 2.569764 | lrm: 0.82 | dt: 1193.58ms | tok/sec: 878,511 | bf16_mfu: 53.96 | total time: 52.15m | eta: 58.8m step 02624/05568 (47.13%) | loss: 2.584375 | lrm: 0.82 | dt: 1185.58ms | tok/sec: 884,439 | bf16_mfu: 54.33 | total time: 52.17m | eta: 58.7m step 02625/05568 (47.14%) | loss: 2.572765 | lrm: 0.82 | dt: 1193.24ms | tok/sec: 878,760 | bf16_mfu: 53.98 | total time: 52.19m | eta: 58.7m step 02626/05568 (47.16%) | loss: 2.569856 | lrm: 0.82 | dt: 1193.62ms | tok/sec: 878,485 | bf16_mfu: 53.96 | total time: 52.21m | eta: 58.7m step 02627/05568 (47.18%) | loss: 2.572561 | lrm: 0.82 | dt: 1189.13ms | tok/sec: 881,798 | bf16_mfu: 54.17 | total time: 52.23m | eta: 58.7m step 02628/05568 (47.20%) | loss: 2.570800 | lrm: 0.82 | dt: 1191.88ms | tok/sec: 879,763 | bf16_mfu: 54.04 | total time: 52.25m | eta: 58.7m step 02629/05568 (47.22%) | loss: 2.574580 | lrm: 0.82 | dt: 1190.83ms | tok/sec: 880,541 | bf16_mfu: 54.09 | total time: 52.27m | eta: 58.6m step 02630/05568 (47.23%) | loss: 2.580904 | lrm: 0.82 | dt: 1193.25ms | tok/sec: 878,752 | bf16_mfu: 53.98 | total time: 52.29m | eta: 58.6m step 02631/05568 (47.25%) | loss: 2.581664 | lrm: 0.82 | dt: 1193.31ms | tok/sec: 878,708 | bf16_mfu: 53.98 | total time: 52.31m | eta: 58.6m step 02632/05568 (47.27%) | loss: 2.582585 | lrm: 0.82 | dt: 1190.76ms | tok/sec: 880,591 | bf16_mfu: 54.09 | total time: 52.33m | eta: 58.6m step 02633/05568 (47.29%) | loss: 2.592494 | lrm: 0.82 | dt: 1193.06ms | tok/sec: 878,892 | bf16_mfu: 53.99 | total time: 52.35m | eta: 58.6m step 02634/05568 (47.31%) | loss: 2.593890 | lrm: 0.82 | dt: 1188.94ms | tok/sec: 881,941 | bf16_mfu: 54.18 | total time: 52.37m | eta: 58.5m step 02635/05568 (47.32%) | loss: 2.594729 | lrm: 0.82 | dt: 1188.57ms | tok/sec: 882,219 | bf16_mfu: 54.19 | total time: 52.39m | eta: 58.5m step 02636/05568 (47.34%) | loss: 2.598760 | lrm: 0.82 | dt: 1193.45ms | tok/sec: 878,611 | bf16_mfu: 53.97 | total time: 52.41m | eta: 58.5m step 02637/05568 (47.36%) | loss: 2.583693 | lrm: 0.82 | dt: 1191.97ms | tok/sec: 879,696 | bf16_mfu: 54.04 | total time: 52.43m | eta: 58.5m step 02638/05568 (47.38%) | loss: 2.586128 | lrm: 0.82 | dt: 1191.58ms | tok/sec: 879,990 | bf16_mfu: 54.06 | total time: 52.45m | eta: 58.5m step 02639/05568 (47.40%) | loss: 2.577447 | lrm: 0.82 | dt: 1193.15ms | tok/sec: 878,827 | bf16_mfu: 53.98 | total time: 52.46m | eta: 58.4m step 02640/05568 (47.41%) | loss: 2.583228 | lrm: 0.82 | dt: 1197.09ms | tok/sec: 875,940 | bf16_mfu: 53.81 | total time: 52.48m | eta: 58.4m step 02641/05568 (47.43%) | loss: 2.584480 | lrm: 0.82 | dt: 1193.64ms | tok/sec: 878,465 | bf16_mfu: 53.96 | total time: 52.50m | eta: 58.4m step 02642/05568 (47.45%) | loss: 2.579900 | lrm: 0.82 | dt: 1186.36ms | tok/sec: 883,857 | bf16_mfu: 54.29 | total time: 52.52m | eta: 58.4m step 02643/05568 (47.47%) | loss: 2.589532 | lrm: 0.82 | dt: 1191.04ms | tok/sec: 880,388 | bf16_mfu: 54.08 | total time: 52.54m | eta: 58.4m step 02644/05568 (47.49%) | loss: 2.586400 | lrm: 0.82 | dt: 1196.47ms | tok/sec: 876,391 | bf16_mfu: 53.83 | total time: 52.56m | eta: 58.3m step 02645/05568 (47.50%) | loss: 2.583442 | lrm: 0.82 | dt: 1194.88ms | tok/sec: 877,558 | bf16_mfu: 53.91 | total time: 52.58m | eta: 58.3m step 02646/05568 (47.52%) | loss: 2.580970 | lrm: 0.82 | dt: 1195.61ms | tok/sec: 877,019 | bf16_mfu: 53.87 | total time: 52.60m | eta: 58.3m step 02647/05568 (47.54%) | loss: 2.571548 | lrm: 0.82 | dt: 1195.06ms | tok/sec: 877,428 | bf16_mfu: 53.90 | total time: 52.62m | eta: 58.3m step 02648/05568 (47.56%) | loss: 2.573608 | lrm: 0.82 | dt: 1190.49ms | tok/sec: 880,789 | bf16_mfu: 54.10 | total time: 52.64m | eta: 58.3m step 02649/05568 (47.58%) | loss: 2.584270 | lrm: 0.82 | dt: 1190.15ms | tok/sec: 881,048 | bf16_mfu: 54.12 | total time: 52.66m | eta: 58.2m step 02650/05568 (47.59%) | loss: 2.596094 | lrm: 0.82 | dt: 1193.62ms | tok/sec: 878,484 | bf16_mfu: 53.96 | total time: 52.68m | eta: 58.2m step 02651/05568 (47.61%) | loss: 2.602271 | lrm: 0.82 | dt: 1193.43ms | tok/sec: 878,624 | bf16_mfu: 53.97 | total time: 52.70m | eta: 58.2m step 02652/05568 (47.63%) | loss: 2.589594 | lrm: 0.82 | dt: 1189.37ms | tok/sec: 881,624 | bf16_mfu: 54.16 | total time: 52.72m | eta: 58.2m step 02653/05568 (47.65%) | loss: 2.593562 | lrm: 0.82 | dt: 1191.22ms | tok/sec: 880,251 | bf16_mfu: 54.07 | total time: 52.74m | eta: 58.2m step 02654/05568 (47.67%) | loss: 2.580772 | lrm: 0.81 | dt: 1189.68ms | tok/sec: 881,396 | bf16_mfu: 54.14 | total time: 52.76m | eta: 58.1m step 02655/05568 (47.68%) | loss: 2.579111 | lrm: 0.81 | dt: 1190.14ms | tok/sec: 881,053 | bf16_mfu: 54.12 | total time: 52.78m | eta: 58.1m step 02656/05568 (47.70%) | loss: 2.588588 | lrm: 0.81 | dt: 1190.86ms | tok/sec: 880,518 | bf16_mfu: 54.09 | total time: 52.80m | eta: 58.1m step 02657/05568 (47.72%) | loss: 2.557839 | lrm: 0.81 | dt: 1191.51ms | tok/sec: 880,040 | bf16_mfu: 54.06 | total time: 52.82m | eta: 58.1m step 02658/05568 (47.74%) | loss: 2.548345 | lrm: 0.81 | dt: 1188.95ms | tok/sec: 881,938 | bf16_mfu: 54.18 | total time: 52.84m | eta: 58.1m step 02659/05568 (47.76%) | loss: 2.549852 | lrm: 0.81 | dt: 1194.06ms | tok/sec: 878,158 | bf16_mfu: 53.94 | total time: 52.86m | eta: 58.0m step 02660/05568 (47.77%) | loss: 2.552299 | lrm: 0.81 | dt: 1190.42ms | tok/sec: 880,844 | bf16_mfu: 54.11 | total time: 52.88m | eta: 58.0m step 02661/05568 (47.79%) | loss: 2.546754 | lrm: 0.81 | dt: 1189.50ms | tok/sec: 881,528 | bf16_mfu: 54.15 | total time: 52.90m | eta: 58.0m step 02662/05568 (47.81%) | loss: 2.549736 | lrm: 0.81 | dt: 1192.74ms | tok/sec: 879,129 | bf16_mfu: 54.00 | total time: 52.92m | eta: 58.0m step 02663/05568 (47.83%) | loss: 2.543437 | lrm: 0.81 | dt: 1197.79ms | tok/sec: 875,422 | bf16_mfu: 53.78 | total time: 52.94m | eta: 58.0m step 02664/05568 (47.84%) | loss: 2.545391 | lrm: 0.81 | dt: 1193.03ms | tok/sec: 878,919 | bf16_mfu: 53.99 | total time: 52.96m | eta: 57.9m step 02665/05568 (47.86%) | loss: 2.551647 | lrm: 0.81 | dt: 1194.62ms | tok/sec: 877,745 | bf16_mfu: 53.92 | total time: 52.98m | eta: 57.9m step 02666/05568 (47.88%) | loss: 2.553213 | lrm: 0.81 | dt: 1190.83ms | tok/sec: 880,540 | bf16_mfu: 54.09 | total time: 53.00m | eta: 57.9m step 02667/05568 (47.90%) | loss: 2.558616 | lrm: 0.81 | dt: 1190.58ms | tok/sec: 880,730 | bf16_mfu: 54.10 | total time: 53.02m | eta: 57.9m step 02668/05568 (47.92%) | loss: 2.564210 | lrm: 0.81 | dt: 1187.38ms | tok/sec: 883,101 | bf16_mfu: 54.25 | total time: 53.04m | eta: 57.9m step 02669/05568 (47.93%) | loss: 2.574900 | lrm: 0.81 | dt: 1190.52ms | tok/sec: 880,771 | bf16_mfu: 54.10 | total time: 53.06m | eta: 57.8m step 02670/05568 (47.95%) | loss: 2.573835 | lrm: 0.81 | dt: 1188.22ms | tok/sec: 882,474 | bf16_mfu: 54.21 | total time: 53.08m | eta: 57.8m step 02671/05568 (47.97%) | loss: 2.573397 | lrm: 0.81 | dt: 1192.92ms | tok/sec: 879,000 | bf16_mfu: 53.99 | total time: 53.10m | eta: 57.8m step 02672/05568 (47.99%) | loss: 2.567888 | lrm: 0.81 | dt: 1189.82ms | tok/sec: 881,290 | bf16_mfu: 54.14 | total time: 53.12m | eta: 57.8m step 02673/05568 (48.01%) | loss: 2.574713 | lrm: 0.81 | dt: 1192.69ms | tok/sec: 879,166 | bf16_mfu: 54.01 | total time: 53.14m | eta: 57.7m step 02674/05568 (48.02%) | loss: 2.575132 | lrm: 0.81 | dt: 1190.54ms | tok/sec: 880,758 | bf16_mfu: 54.10 | total time: 53.16m | eta: 57.7m step 02675/05568 (48.04%) | loss: 2.587461 | lrm: 0.81 | dt: 1196.01ms | tok/sec: 876,728 | bf16_mfu: 53.86 | total time: 53.18m | eta: 57.7m step 02676/05568 (48.06%) | loss: 2.587511 | lrm: 0.81 | dt: 1192.86ms | tok/sec: 879,041 | bf16_mfu: 54.00 | total time: 53.20m | eta: 57.7m step 02677/05568 (48.08%) | loss: 2.586108 | lrm: 0.81 | dt: 1192.58ms | tok/sec: 879,250 | bf16_mfu: 54.01 | total time: 53.22m | eta: 57.7m step 02678/05568 (48.10%) | loss: 2.592537 | lrm: 0.81 | dt: 1192.70ms | tok/sec: 879,165 | bf16_mfu: 54.01 | total time: 53.24m | eta: 57.6m step 02679/05568 (48.11%) | loss: 2.590903 | lrm: 0.81 | dt: 1192.75ms | tok/sec: 879,124 | bf16_mfu: 54.00 | total time: 53.26m | eta: 57.6m step 02680/05568 (48.13%) | loss: 2.594109 | lrm: 0.81 | dt: 1189.21ms | tok/sec: 881,742 | bf16_mfu: 54.16 | total time: 53.28m | eta: 57.6m step 02681/05568 (48.15%) | loss: 2.588761 | lrm: 0.81 | dt: 1194.67ms | tok/sec: 877,714 | bf16_mfu: 53.92 | total time: 53.30m | eta: 57.6m step 02682/05568 (48.17%) | loss: 2.582389 | lrm: 0.81 | dt: 1191.27ms | tok/sec: 880,214 | bf16_mfu: 54.07 | total time: 53.32m | eta: 57.6m step 02683/05568 (48.19%) | loss: 2.582271 | lrm: 0.81 | dt: 1191.33ms | tok/sec: 880,172 | bf16_mfu: 54.07 | total time: 53.34m | eta: 57.5m step 02684/05568 (48.20%) | loss: 2.578316 | lrm: 0.81 | dt: 1187.45ms | tok/sec: 883,045 | bf16_mfu: 54.24 | total time: 53.36m | eta: 57.5m step 02685/05568 (48.22%) | loss: 2.566410 | lrm: 0.81 | dt: 1194.64ms | tok/sec: 877,736 | bf16_mfu: 53.92 | total time: 53.38m | eta: 57.5m step 02686/05568 (48.24%) | loss: 2.558017 | lrm: 0.81 | dt: 1194.93ms | tok/sec: 877,521 | bf16_mfu: 53.90 | total time: 53.40m | eta: 57.5m step 02687/05568 (48.26%) | loss: 2.559138 | lrm: 0.81 | dt: 1191.17ms | tok/sec: 880,293 | bf16_mfu: 54.07 | total time: 53.42m | eta: 57.5m step 02688/05568 (48.28%) | loss: 2.556855 | lrm: 0.81 | dt: 1194.97ms | tok/sec: 877,492 | bf16_mfu: 53.90 | total time: 53.44m | eta: 57.4m step 02689/05568 (48.29%) | loss: 2.560588 | lrm: 0.81 | dt: 1196.05ms | tok/sec: 876,697 | bf16_mfu: 53.85 | total time: 53.46m | eta: 57.4m step 02690/05568 (48.31%) | loss: 2.566104 | lrm: 0.81 | dt: 1195.89ms | tok/sec: 876,819 | bf16_mfu: 53.86 | total time: 53.48m | eta: 57.4m step 02691/05568 (48.33%) | loss: 2.582075 | lrm: 0.81 | dt: 1193.50ms | tok/sec: 878,575 | bf16_mfu: 53.97 | total time: 53.50m | eta: 57.4m step 02692/05568 (48.35%) | loss: 2.563317 | lrm: 0.80 | dt: 1189.56ms | tok/sec: 881,483 | bf16_mfu: 54.15 | total time: 53.52m | eta: 57.4m step 02693/05568 (48.37%) | loss: 2.569689 | lrm: 0.80 | dt: 1193.55ms | tok/sec: 878,538 | bf16_mfu: 53.97 | total time: 53.54m | eta: 57.3m step 02694/05568 (48.38%) | loss: 2.577363 | lrm: 0.80 | dt: 1188.26ms | tok/sec: 882,446 | bf16_mfu: 54.21 | total time: 53.56m | eta: 57.3m step 02695/05568 (48.40%) | loss: 2.575186 | lrm: 0.80 | dt: 1189.78ms | tok/sec: 881,319 | bf16_mfu: 54.14 | total time: 53.58m | eta: 57.3m step 02696/05568 (48.42%) | loss: 2.585741 | lrm: 0.80 | dt: 1187.72ms | tok/sec: 882,847 | bf16_mfu: 54.23 | total time: 53.60m | eta: 57.3m step 02697/05568 (48.44%) | loss: 2.581137 | lrm: 0.80 | dt: 1191.70ms | tok/sec: 879,896 | bf16_mfu: 54.05 | total time: 53.62m | eta: 57.3m step 02698/05568 (48.46%) | loss: 2.572791 | lrm: 0.80 | dt: 1189.55ms | tok/sec: 881,487 | bf16_mfu: 54.15 | total time: 53.64m | eta: 57.2m step 02699/05568 (48.47%) | loss: 2.574480 | lrm: 0.80 | dt: 1194.33ms | tok/sec: 877,962 | bf16_mfu: 53.93 | total time: 53.66m | eta: 57.2m step 02700/05568 (48.49%) | loss: 2.569306 | lrm: 0.80 | dt: 1194.34ms | tok/sec: 877,954 | bf16_mfu: 53.93 | total time: 53.68m | eta: 57.2m step 02701/05568 (48.51%) | loss: 2.574541 | lrm: 0.80 | dt: 1189.44ms | tok/sec: 881,573 | bf16_mfu: 54.15 | total time: 53.70m | eta: 57.2m step 02702/05568 (48.53%) | loss: 2.557599 | lrm: 0.80 | dt: 1189.61ms | tok/sec: 881,446 | bf16_mfu: 54.15 | total time: 53.72m | eta: 57.2m step 02703/05568 (48.55%) | loss: 2.552844 | lrm: 0.80 | dt: 1188.48ms | tok/sec: 882,279 | bf16_mfu: 54.20 | total time: 53.74m | eta: 57.1m step 02704/05568 (48.56%) | loss: 2.564667 | lrm: 0.80 | dt: 1189.88ms | tok/sec: 881,243 | bf16_mfu: 54.13 | total time: 53.76m | eta: 57.1m step 02705/05568 (48.58%) | loss: 2.557783 | lrm: 0.80 | dt: 1194.38ms | tok/sec: 877,924 | bf16_mfu: 53.93 | total time: 53.78m | eta: 57.1m step 02706/05568 (48.60%) | loss: 2.557386 | lrm: 0.80 | dt: 1189.39ms | tok/sec: 881,610 | bf16_mfu: 54.16 | total time: 53.80m | eta: 57.1m step 02707/05568 (48.62%) | loss: 2.555226 | lrm: 0.80 | dt: 1190.67ms | tok/sec: 880,658 | bf16_mfu: 54.10 | total time: 53.82m | eta: 57.1m step 02708/05568 (48.64%) | loss: 2.570372 | lrm: 0.80 | dt: 1189.76ms | tok/sec: 881,335 | bf16_mfu: 54.14 | total time: 53.84m | eta: 57.0m step 02709/05568 (48.65%) | loss: 2.577005 | lrm: 0.80 | dt: 1192.25ms | tok/sec: 879,490 | bf16_mfu: 54.03 | total time: 53.86m | eta: 57.0m step 02710/05568 (48.67%) | loss: 2.578389 | lrm: 0.80 | dt: 1191.16ms | tok/sec: 880,297 | bf16_mfu: 54.07 | total time: 53.88m | eta: 57.0m step 02711/05568 (48.69%) | loss: 2.577781 | lrm: 0.80 | dt: 1193.24ms | tok/sec: 878,765 | bf16_mfu: 53.98 | total time: 53.90m | eta: 57.0m step 02712/05568 (48.71%) | loss: 2.566396 | lrm: 0.80 | dt: 1198.57ms | tok/sec: 874,857 | bf16_mfu: 53.74 | total time: 53.92m | eta: 57.0m step 02713/05568 (48.72%) | loss: 2.570524 | lrm: 0.80 | dt: 1189.85ms | tok/sec: 881,267 | bf16_mfu: 54.13 | total time: 53.93m | eta: 56.9m step 02714/05568 (48.74%) | loss: 2.576234 | lrm: 0.80 | dt: 1193.48ms | tok/sec: 878,586 | bf16_mfu: 53.97 | total time: 53.95m | eta: 56.9m step 02715/05568 (48.76%) | loss: 2.577097 | lrm: 0.80 | dt: 1196.20ms | tok/sec: 876,588 | bf16_mfu: 53.85 | total time: 53.97m | eta: 56.9m step 02716/05568 (48.78%) | loss: 2.570247 | lrm: 0.80 | dt: 1189.82ms | tok/sec: 881,291 | bf16_mfu: 54.14 | total time: 53.99m | eta: 56.9m step 02717/05568 (48.80%) | loss: 2.565740 | lrm: 0.80 | dt: 1195.95ms | tok/sec: 876,770 | bf16_mfu: 53.86 | total time: 54.01m | eta: 56.9m step 02718/05568 (48.81%) | loss: 2.563593 | lrm: 0.80 | dt: 1186.70ms | tok/sec: 883,608 | bf16_mfu: 54.28 | total time: 54.03m | eta: 56.8m step 02719/05568 (48.83%) | loss: 2.561530 | lrm: 0.80 | dt: 1192.24ms | tok/sec: 879,502 | bf16_mfu: 54.03 | total time: 54.05m | eta: 56.8m step 02720/05568 (48.85%) | loss: 2.556497 | lrm: 0.80 | dt: 1190.70ms | tok/sec: 880,640 | bf16_mfu: 54.10 | total time: 54.07m | eta: 56.8m step 02721/05568 (48.87%) | loss: 2.554554 | lrm: 0.80 | dt: 1190.79ms | tok/sec: 880,573 | bf16_mfu: 54.09 | total time: 54.09m | eta: 56.8m step 02722/05568 (48.89%) | loss: 2.549167 | lrm: 0.80 | dt: 1190.62ms | tok/sec: 880,695 | bf16_mfu: 54.10 | total time: 54.11m | eta: 56.8m step 02723/05568 (48.90%) | loss: 2.547499 | lrm: 0.80 | dt: 1196.29ms | tok/sec: 876,525 | bf16_mfu: 53.84 | total time: 54.13m | eta: 56.7m step 02724/05568 (48.92%) | loss: 2.549938 | lrm: 0.80 | dt: 1189.54ms | tok/sec: 881,497 | bf16_mfu: 54.15 | total time: 54.15m | eta: 56.7m step 02725/05568 (48.94%) | loss: 2.550366 | lrm: 0.80 | dt: 1190.57ms | tok/sec: 880,733 | bf16_mfu: 54.10 | total time: 54.17m | eta: 56.7m step 02726/05568 (48.96%) | loss: 2.559695 | lrm: 0.80 | dt: 1190.97ms | tok/sec: 880,436 | bf16_mfu: 54.08 | total time: 54.19m | eta: 56.7m step 02727/05568 (48.98%) | loss: 2.565311 | lrm: 0.80 | dt: 1191.91ms | tok/sec: 879,743 | bf16_mfu: 54.04 | total time: 54.21m | eta: 56.7m step 02728/05568 (48.99%) | loss: 2.569282 | lrm: 0.80 | dt: 1194.25ms | tok/sec: 878,020 | bf16_mfu: 53.93 | total time: 54.23m | eta: 56.6m step 02729/05568 (49.01%) | loss: 2.568716 | lrm: 0.80 | dt: 1190.97ms | tok/sec: 880,441 | bf16_mfu: 54.08 | total time: 54.25m | eta: 56.6m step 02730/05568 (49.03%) | loss: 2.570174 | lrm: 0.79 | dt: 1188.85ms | tok/sec: 882,008 | bf16_mfu: 54.18 | total time: 54.27m | eta: 56.6m step 02731/05568 (49.05%) | loss: 2.578558 | lrm: 0.79 | dt: 1195.96ms | tok/sec: 876,766 | bf16_mfu: 53.86 | total time: 54.29m | eta: 56.6m step 02732/05568 (49.07%) | loss: 2.570453 | lrm: 0.79 | dt: 1196.16ms | tok/sec: 876,618 | bf16_mfu: 53.85 | total time: 54.31m | eta: 56.6m step 02733/05568 (49.08%) | loss: 2.576747 | lrm: 0.79 | dt: 1197.17ms | tok/sec: 875,880 | bf16_mfu: 53.80 | total time: 54.33m | eta: 56.5m step 02734/05568 (49.10%) | loss: 2.579434 | lrm: 0.79 | dt: 1190.88ms | tok/sec: 880,506 | bf16_mfu: 54.09 | total time: 54.35m | eta: 56.5m step 02735/05568 (49.12%) | loss: 2.568331 | lrm: 0.79 | dt: 1188.72ms | tok/sec: 882,107 | bf16_mfu: 54.19 | total time: 54.37m | eta: 56.5m step 02736/05568 (49.14%) | loss: 2.564242 | lrm: 0.79 | dt: 1188.82ms | tok/sec: 882,031 | bf16_mfu: 54.18 | total time: 54.39m | eta: 56.5m step 02737/05568 (49.16%) | loss: 2.564413 | lrm: 0.79 | dt: 1192.52ms | tok/sec: 879,290 | bf16_mfu: 54.01 | total time: 54.41m | eta: 56.5m step 02738/05568 (49.17%) | loss: 2.556944 | lrm: 0.79 | dt: 1189.72ms | tok/sec: 881,360 | bf16_mfu: 54.14 | total time: 54.43m | eta: 56.4m step 02739/05568 (49.19%) | loss: 2.561070 | lrm: 0.79 | dt: 1194.70ms | tok/sec: 877,686 | bf16_mfu: 53.91 | total time: 54.45m | eta: 56.4m step 02740/05568 (49.21%) | loss: 2.569398 | lrm: 0.79 | dt: 1192.51ms | tok/sec: 879,304 | bf16_mfu: 54.01 | total time: 54.47m | eta: 56.4m step 02741/05568 (49.23%) | loss: 2.562734 | lrm: 0.79 | dt: 1187.37ms | tok/sec: 883,105 | bf16_mfu: 54.25 | total time: 54.49m | eta: 56.4m step 02742/05568 (49.25%) | loss: 2.563789 | lrm: 0.79 | dt: 1188.08ms | tok/sec: 882,577 | bf16_mfu: 54.21 | total time: 54.51m | eta: 56.4m step 02743/05568 (49.26%) | loss: 2.570458 | lrm: 0.79 | dt: 1195.85ms | tok/sec: 876,842 | bf16_mfu: 53.86 | total time: 54.53m | eta: 56.3m step 02744/05568 (49.28%) | loss: 2.563364 | lrm: 0.79 | dt: 1189.67ms | tok/sec: 881,400 | bf16_mfu: 54.14 | total time: 54.55m | eta: 56.3m step 02745/05568 (49.30%) | loss: 2.568037 | lrm: 0.79 | dt: 1193.28ms | tok/sec: 878,737 | bf16_mfu: 53.98 | total time: 54.57m | eta: 56.3m step 02746/05568 (49.32%) | loss: 2.580570 | lrm: 0.79 | dt: 1194.25ms | tok/sec: 878,020 | bf16_mfu: 53.93 | total time: 54.59m | eta: 56.3m step 02747/05568 (49.34%) | loss: 2.591058 | lrm: 0.79 | dt: 1195.55ms | tok/sec: 877,063 | bf16_mfu: 53.88 | total time: 54.61m | eta: 56.3m step 02748/05568 (49.35%) | loss: 2.586473 | lrm: 0.79 | dt: 1189.95ms | tok/sec: 881,192 | bf16_mfu: 54.13 | total time: 54.63m | eta: 56.2m step 02749/05568 (49.37%) | loss: 2.582253 | lrm: 0.79 | dt: 1188.53ms | tok/sec: 882,244 | bf16_mfu: 54.19 | total time: 54.65m | eta: 56.2m step:2750/5568 val_bpb:0.781275 val_time:3.89s step 02750/05568 (49.39%) | loss: 2.572410 | lrm: 0.79 | dt: 1189.01ms | tok/sec: 881,889 | bf16_mfu: 54.17 | total time: 54.67m | eta: 56.2m step 02751/05568 (49.41%) | loss: 2.569667 | lrm: 0.79 | dt: 1194.23ms | tok/sec: 878,037 | bf16_mfu: 53.94 | total time: 54.69m | eta: 56.2m step 02752/05568 (49.43%) | loss: 2.556464 | lrm: 0.79 | dt: 1192.09ms | tok/sec: 879,608 | bf16_mfu: 54.03 | total time: 54.71m | eta: 56.2m step 02753/05568 (49.44%) | loss: 2.562674 | lrm: 0.79 | dt: 1192.43ms | tok/sec: 879,362 | bf16_mfu: 54.02 | total time: 54.73m | eta: 56.1m step 02754/05568 (49.46%) | loss: 2.565242 | lrm: 0.79 | dt: 1194.73ms | tok/sec: 877,671 | bf16_mfu: 53.91 | total time: 54.75m | eta: 56.1m step 02755/05568 (49.48%) | loss: 2.563353 | lrm: 0.79 | dt: 1196.53ms | tok/sec: 876,345 | bf16_mfu: 53.83 | total time: 54.77m | eta: 56.1m step 02756/05568 (49.50%) | loss: 2.567423 | lrm: 0.79 | dt: 1192.51ms | tok/sec: 879,303 | bf16_mfu: 54.01 | total time: 54.79m | eta: 56.1m step 02757/05568 (49.52%) | loss: 2.549569 | lrm: 0.79 | dt: 1192.34ms | tok/sec: 879,424 | bf16_mfu: 54.02 | total time: 54.81m | eta: 56.1m step 02758/05568 (49.53%) | loss: 2.541260 | lrm: 0.79 | dt: 1192.56ms | tok/sec: 879,261 | bf16_mfu: 54.01 | total time: 54.83m | eta: 56.0m step 02759/05568 (49.55%) | loss: 2.550341 | lrm: 0.79 | dt: 1198.92ms | tok/sec: 874,602 | bf16_mfu: 53.72 | total time: 54.85m | eta: 56.0m step 02760/05568 (49.57%) | loss: 2.559807 | lrm: 0.79 | dt: 1196.86ms | tok/sec: 876,104 | bf16_mfu: 53.82 | total time: 54.87m | eta: 56.0m step 02761/05568 (49.59%) | loss: 2.563313 | lrm: 0.79 | dt: 1191.93ms | tok/sec: 879,729 | bf16_mfu: 54.04 | total time: 54.89m | eta: 56.0m step 02762/05568 (49.60%) | loss: 2.555598 | lrm: 0.79 | dt: 1194.32ms | tok/sec: 877,968 | bf16_mfu: 53.93 | total time: 54.91m | eta: 56.0m step 02763/05568 (49.62%) | loss: 2.554426 | lrm: 0.79 | dt: 1190.52ms | tok/sec: 880,770 | bf16_mfu: 54.10 | total time: 54.93m | eta: 55.9m step 02764/05568 (49.64%) | loss: 2.548280 | lrm: 0.79 | dt: 1191.93ms | tok/sec: 879,729 | bf16_mfu: 54.04 | total time: 54.95m | eta: 55.9m step 02765/05568 (49.66%) | loss: 2.553538 | lrm: 0.79 | dt: 1193.89ms | tok/sec: 878,284 | bf16_mfu: 53.95 | total time: 54.97m | eta: 55.9m step 02766/05568 (49.68%) | loss: 2.551846 | lrm: 0.79 | dt: 1194.28ms | tok/sec: 878,000 | bf16_mfu: 53.93 | total time: 54.99m | eta: 55.9m step 02767/05568 (49.69%) | loss: 2.560168 | lrm: 0.79 | dt: 1195.07ms | tok/sec: 877,421 | bf16_mfu: 53.90 | total time: 55.01m | eta: 55.9m step 02768/05568 (49.71%) | loss: 2.556814 | lrm: 0.79 | dt: 1192.40ms | tok/sec: 879,384 | bf16_mfu: 54.02 | total time: 55.03m | eta: 55.8m step 02769/05568 (49.73%) | loss: 2.557946 | lrm: 0.78 | dt: 1193.44ms | tok/sec: 878,614 | bf16_mfu: 53.97 | total time: 55.05m | eta: 55.8m step 02770/05568 (49.75%) | loss: 2.561625 | lrm: 0.78 | dt: 1194.48ms | tok/sec: 877,850 | bf16_mfu: 53.92 | total time: 55.07m | eta: 55.8m step 02771/05568 (49.77%) | loss: 2.572857 | lrm: 0.78 | dt: 1189.06ms | tok/sec: 881,850 | bf16_mfu: 54.17 | total time: 55.09m | eta: 55.8m step 02772/05568 (49.78%) | loss: 2.560900 | lrm: 0.78 | dt: 1195.13ms | tok/sec: 877,370 | bf16_mfu: 53.89 | total time: 55.11m | eta: 55.8m step 02773/05568 (49.80%) | loss: 2.559355 | lrm: 0.78 | dt: 1197.04ms | tok/sec: 875,970 | bf16_mfu: 53.81 | total time: 55.13m | eta: 55.7m step 02774/05568 (49.82%) | loss: 2.578424 | lrm: 0.78 | dt: 1193.85ms | tok/sec: 878,314 | bf16_mfu: 53.95 | total time: 55.15m | eta: 55.7m step 02775/05568 (49.84%) | loss: 2.573980 | lrm: 0.78 | dt: 1197.61ms | tok/sec: 875,557 | bf16_mfu: 53.78 | total time: 55.17m | eta: 55.7m step 02776/05568 (49.86%) | loss: 2.564088 | lrm: 0.78 | dt: 1195.84ms | tok/sec: 876,855 | bf16_mfu: 53.86 | total time: 55.19m | eta: 55.7m step 02777/05568 (49.87%) | loss: 2.570452 | lrm: 0.78 | dt: 1187.95ms | tok/sec: 882,674 | bf16_mfu: 54.22 | total time: 55.21m | eta: 55.7m step 02778/05568 (49.89%) | loss: 2.563712 | lrm: 0.78 | dt: 1192.93ms | tok/sec: 878,995 | bf16_mfu: 53.99 | total time: 55.23m | eta: 55.6m step 02779/05568 (49.91%) | loss: 2.565110 | lrm: 0.78 | dt: 1194.08ms | tok/sec: 878,142 | bf16_mfu: 53.94 | total time: 55.25m | eta: 55.6m step 02780/05568 (49.93%) | loss: 2.565119 | lrm: 0.78 | dt: 1189.59ms | tok/sec: 881,463 | bf16_mfu: 54.15 | total time: 55.27m | eta: 55.6m step 02781/05568 (49.95%) | loss: 2.571014 | lrm: 0.78 | dt: 1193.77ms | tok/sec: 878,375 | bf16_mfu: 53.96 | total time: 55.29m | eta: 55.6m step 02782/05568 (49.96%) | loss: 2.570715 | lrm: 0.78 | dt: 1194.74ms | tok/sec: 877,657 | bf16_mfu: 53.91 | total time: 55.31m | eta: 55.6m step 02783/05568 (49.98%) | loss: 2.566263 | lrm: 0.78 | dt: 1194.22ms | tok/sec: 878,041 | bf16_mfu: 53.94 | total time: 55.33m | eta: 55.5m step 02784/05568 (50.00%) | loss: 2.574717 | lrm: 0.78 | dt: 1195.23ms | tok/sec: 877,297 | bf16_mfu: 53.89 | total time: 55.35m | eta: 55.5m step 02785/05568 (50.02%) | loss: 2.582091 | lrm: 0.78 | dt: 1194.07ms | tok/sec: 878,151 | bf16_mfu: 53.94 | total time: 55.37m | eta: 55.5m step 02786/05568 (50.04%) | loss: 2.584202 | lrm: 0.78 | dt: 1192.82ms | tok/sec: 879,073 | bf16_mfu: 54.00 | total time: 55.39m | eta: 55.5m step 02787/05568 (50.05%) | loss: 2.589888 | lrm: 0.78 | dt: 1195.79ms | tok/sec: 876,886 | bf16_mfu: 53.87 | total time: 55.41m | eta: 55.5m step 02788/05568 (50.07%) | loss: 2.588501 | lrm: 0.78 | dt: 1190.70ms | tok/sec: 880,639 | bf16_mfu: 54.10 | total time: 55.43m | eta: 55.4m step 02789/05568 (50.09%) | loss: 2.599519 | lrm: 0.78 | dt: 1191.62ms | tok/sec: 879,960 | bf16_mfu: 54.05 | total time: 55.45m | eta: 55.4m step 02790/05568 (50.11%) | loss: 2.593760 | lrm: 0.78 | dt: 1197.04ms | tok/sec: 875,974 | bf16_mfu: 53.81 | total time: 55.47m | eta: 55.4m step 02791/05568 (50.13%) | loss: 2.592130 | lrm: 0.78 | dt: 1198.10ms | tok/sec: 875,200 | bf16_mfu: 53.76 | total time: 55.49m | eta: 55.4m step 02792/05568 (50.14%) | loss: 2.589692 | lrm: 0.78 | dt: 1197.14ms | tok/sec: 875,899 | bf16_mfu: 53.80 | total time: 55.51m | eta: 55.4m step 02793/05568 (50.16%) | loss: 2.579622 | lrm: 0.78 | dt: 1192.35ms | tok/sec: 879,420 | bf16_mfu: 54.02 | total time: 55.53m | eta: 55.3m step 02794/05568 (50.18%) | loss: 2.569440 | lrm: 0.78 | dt: 1189.27ms | tok/sec: 881,697 | bf16_mfu: 54.16 | total time: 55.55m | eta: 55.3m step 02795/05568 (50.20%) | loss: 2.572852 | lrm: 0.78 | dt: 1190.94ms | tok/sec: 880,460 | bf16_mfu: 54.08 | total time: 55.57m | eta: 55.3m step 02796/05568 (50.22%) | loss: 2.565604 | lrm: 0.78 | dt: 1191.78ms | tok/sec: 879,842 | bf16_mfu: 54.05 | total time: 55.59m | eta: 55.3m step 02797/05568 (50.23%) | loss: 2.575788 | lrm: 0.78 | dt: 1190.95ms | tok/sec: 880,455 | bf16_mfu: 54.08 | total time: 55.60m | eta: 55.3m step 02798/05568 (50.25%) | loss: 2.579270 | lrm: 0.78 | dt: 1196.66ms | tok/sec: 876,251 | bf16_mfu: 53.83 | total time: 55.62m | eta: 55.2m step 02799/05568 (50.27%) | loss: 2.568164 | lrm: 0.78 | dt: 1194.17ms | tok/sec: 878,078 | bf16_mfu: 53.94 | total time: 55.64m | eta: 55.2m step 02800/05568 (50.29%) | loss: 2.569659 | lrm: 0.78 | dt: 1190.30ms | tok/sec: 880,934 | bf16_mfu: 54.11 | total time: 55.66m | eta: 55.2m step 02801/05568 (50.31%) | loss: 2.559441 | lrm: 0.78 | dt: 1191.46ms | tok/sec: 880,075 | bf16_mfu: 54.06 | total time: 55.68m | eta: 55.2m step 02802/05568 (50.32%) | loss: 2.550785 | lrm: 0.78 | dt: 1194.36ms | tok/sec: 877,942 | bf16_mfu: 53.93 | total time: 55.70m | eta: 55.2m step 02803/05568 (50.34%) | loss: 2.545212 | lrm: 0.78 | dt: 1195.16ms | tok/sec: 877,355 | bf16_mfu: 53.89 | total time: 55.72m | eta: 55.1m step 02804/05568 (50.36%) | loss: 2.552409 | lrm: 0.78 | dt: 1193.84ms | tok/sec: 878,321 | bf16_mfu: 53.95 | total time: 55.74m | eta: 55.1m step 02805/05568 (50.38%) | loss: 2.552733 | lrm: 0.78 | dt: 1193.45ms | tok/sec: 878,607 | bf16_mfu: 53.97 | total time: 55.76m | eta: 55.1m step 02806/05568 (50.40%) | loss: 2.553450 | lrm: 0.78 | dt: 1192.52ms | tok/sec: 879,292 | bf16_mfu: 54.01 | total time: 55.78m | eta: 55.1m step 02807/05568 (50.41%) | loss: 2.549598 | lrm: 0.77 | dt: 1196.18ms | tok/sec: 876,602 | bf16_mfu: 53.85 | total time: 55.80m | eta: 55.1m step 02808/05568 (50.43%) | loss: 2.547012 | lrm: 0.77 | dt: 1195.08ms | tok/sec: 877,409 | bf16_mfu: 53.90 | total time: 55.82m | eta: 55.0m step 02809/05568 (50.45%) | loss: 2.553698 | lrm: 0.77 | dt: 1192.88ms | tok/sec: 879,026 | bf16_mfu: 54.00 | total time: 55.84m | eta: 55.0m step 02810/05568 (50.47%) | loss: 2.558533 | lrm: 0.77 | dt: 1194.28ms | tok/sec: 877,997 | bf16_mfu: 53.93 | total time: 55.86m | eta: 55.0m step 02811/05568 (50.48%) | loss: 2.563310 | lrm: 0.77 | dt: 1190.11ms | tok/sec: 881,078 | bf16_mfu: 54.12 | total time: 55.88m | eta: 55.0m step 02812/05568 (50.50%) | loss: 2.566731 | lrm: 0.77 | dt: 1192.03ms | tok/sec: 879,657 | bf16_mfu: 54.04 | total time: 55.90m | eta: 55.0m step 02813/05568 (50.52%) | loss: 2.565822 | lrm: 0.77 | dt: 1194.57ms | tok/sec: 877,782 | bf16_mfu: 53.92 | total time: 55.92m | eta: 54.9m step 02814/05568 (50.54%) | loss: 2.562214 | lrm: 0.77 | dt: 1188.42ms | tok/sec: 882,330 | bf16_mfu: 54.20 | total time: 55.94m | eta: 54.9m step 02815/05568 (50.56%) | loss: 2.557540 | lrm: 0.77 | dt: 1196.34ms | tok/sec: 876,485 | bf16_mfu: 53.84 | total time: 55.96m | eta: 54.9m step 02816/05568 (50.57%) | loss: 2.567655 | lrm: 0.77 | dt: 1194.96ms | tok/sec: 877,495 | bf16_mfu: 53.90 | total time: 55.98m | eta: 54.9m step 02817/05568 (50.59%) | loss: 2.569559 | lrm: 0.77 | dt: 1192.44ms | tok/sec: 879,356 | bf16_mfu: 54.02 | total time: 56.00m | eta: 54.9m step 02818/05568 (50.61%) | loss: 2.576082 | lrm: 0.77 | dt: 1197.14ms | tok/sec: 875,901 | bf16_mfu: 53.80 | total time: 56.02m | eta: 54.8m step 02819/05568 (50.63%) | loss: 2.578349 | lrm: 0.77 | dt: 1195.91ms | tok/sec: 876,799 | bf16_mfu: 53.86 | total time: 56.04m | eta: 54.8m step 02820/05568 (50.65%) | loss: 2.566521 | lrm: 0.77 | dt: 1189.92ms | tok/sec: 881,212 | bf16_mfu: 54.13 | total time: 56.06m | eta: 54.8m step 02821/05568 (50.66%) | loss: 2.548625 | lrm: 0.77 | dt: 1194.52ms | tok/sec: 877,820 | bf16_mfu: 53.92 | total time: 56.08m | eta: 54.8m step 02822/05568 (50.68%) | loss: 2.544515 | lrm: 0.77 | dt: 1192.01ms | tok/sec: 879,671 | bf16_mfu: 54.04 | total time: 56.10m | eta: 54.8m step 02823/05568 (50.70%) | loss: 2.548754 | lrm: 0.77 | dt: 1192.84ms | tok/sec: 879,059 | bf16_mfu: 54.00 | total time: 56.12m | eta: 54.7m step 02824/05568 (50.72%) | loss: 2.552561 | lrm: 0.77 | dt: 1194.33ms | tok/sec: 877,962 | bf16_mfu: 53.93 | total time: 56.14m | eta: 54.7m step 02825/05568 (50.74%) | loss: 2.562936 | lrm: 0.77 | dt: 1192.97ms | tok/sec: 878,965 | bf16_mfu: 53.99 | total time: 56.16m | eta: 54.7m step 02826/05568 (50.75%) | loss: 2.563982 | lrm: 0.77 | dt: 1194.65ms | tok/sec: 877,729 | bf16_mfu: 53.92 | total time: 56.18m | eta: 54.7m step 02827/05568 (50.77%) | loss: 2.559514 | lrm: 0.77 | dt: 1194.91ms | tok/sec: 877,533 | bf16_mfu: 53.90 | total time: 56.20m | eta: 54.7m step 02828/05568 (50.79%) | loss: 2.552862 | lrm: 0.77 | dt: 1191.81ms | tok/sec: 879,821 | bf16_mfu: 54.05 | total time: 56.22m | eta: 54.6m step 02829/05568 (50.81%) | loss: 2.557119 | lrm: 0.77 | dt: 1197.03ms | tok/sec: 875,980 | bf16_mfu: 53.81 | total time: 56.24m | eta: 54.6m step 02830/05568 (50.83%) | loss: 2.568241 | lrm: 0.77 | dt: 1193.32ms | tok/sec: 878,706 | bf16_mfu: 53.98 | total time: 56.26m | eta: 54.6m step 02831/05568 (50.84%) | loss: 2.571518 | lrm: 0.77 | dt: 1191.50ms | tok/sec: 880,044 | bf16_mfu: 54.06 | total time: 56.28m | eta: 54.6m step 02832/05568 (50.86%) | loss: 2.574947 | lrm: 0.77 | dt: 1190.71ms | tok/sec: 880,628 | bf16_mfu: 54.09 | total time: 56.30m | eta: 54.6m step 02833/05568 (50.88%) | loss: 2.575117 | lrm: 0.77 | dt: 1189.71ms | tok/sec: 881,367 | bf16_mfu: 54.14 | total time: 56.32m | eta: 54.5m step 02834/05568 (50.90%) | loss: 2.575758 | lrm: 0.77 | dt: 1195.22ms | tok/sec: 877,306 | bf16_mfu: 53.89 | total time: 56.34m | eta: 54.5m step 02835/05568 (50.92%) | loss: 2.573331 | lrm: 0.77 | dt: 1187.95ms | tok/sec: 882,673 | bf16_mfu: 54.22 | total time: 56.36m | eta: 54.5m step 02836/05568 (50.93%) | loss: 2.578179 | lrm: 0.77 | dt: 1197.02ms | tok/sec: 875,986 | bf16_mfu: 53.81 | total time: 56.38m | eta: 54.5m step 02837/05568 (50.95%) | loss: 2.576062 | lrm: 0.77 | dt: 1193.41ms | tok/sec: 878,636 | bf16_mfu: 53.97 | total time: 56.40m | eta: 54.5m step 02838/05568 (50.97%) | loss: 2.565213 | lrm: 0.77 | dt: 1194.47ms | tok/sec: 877,855 | bf16_mfu: 53.92 | total time: 56.42m | eta: 54.4m step 02839/05568 (50.99%) | loss: 2.564143 | lrm: 0.77 | dt: 1198.09ms | tok/sec: 875,206 | bf16_mfu: 53.76 | total time: 56.44m | eta: 54.4m step 02840/05568 (51.01%) | loss: 2.546568 | lrm: 0.77 | dt: 1193.99ms | tok/sec: 878,209 | bf16_mfu: 53.95 | total time: 56.46m | eta: 54.4m step 02841/05568 (51.02%) | loss: 2.547157 | lrm: 0.77 | dt: 1197.05ms | tok/sec: 875,963 | bf16_mfu: 53.81 | total time: 56.48m | eta: 54.4m step 02842/05568 (51.04%) | loss: 2.554371 | lrm: 0.77 | dt: 1197.09ms | tok/sec: 875,940 | bf16_mfu: 53.81 | total time: 56.50m | eta: 54.4m step 02843/05568 (51.06%) | loss: 2.554877 | lrm: 0.77 | dt: 1191.65ms | tok/sec: 879,934 | bf16_mfu: 54.05 | total time: 56.52m | eta: 54.3m step 02844/05568 (51.08%) | loss: 2.554353 | lrm: 0.77 | dt: 1195.27ms | tok/sec: 877,268 | bf16_mfu: 53.89 | total time: 56.54m | eta: 54.3m step 02845/05568 (51.10%) | loss: 2.556588 | lrm: 0.76 | dt: 1188.70ms | tok/sec: 882,116 | bf16_mfu: 54.19 | total time: 56.56m | eta: 54.3m step 02846/05568 (51.11%) | loss: 2.545313 | lrm: 0.76 | dt: 1195.54ms | tok/sec: 877,075 | bf16_mfu: 53.88 | total time: 56.58m | eta: 54.3m step 02847/05568 (51.13%) | loss: 2.550199 | lrm: 0.76 | dt: 1193.63ms | tok/sec: 878,474 | bf16_mfu: 53.96 | total time: 56.60m | eta: 54.3m step 02848/05568 (51.15%) | loss: 2.558478 | lrm: 0.76 | dt: 1197.30ms | tok/sec: 875,785 | bf16_mfu: 53.80 | total time: 56.62m | eta: 54.2m step 02849/05568 (51.17%) | loss: 2.571427 | lrm: 0.76 | dt: 1189.60ms | tok/sec: 881,454 | bf16_mfu: 54.15 | total time: 56.64m | eta: 54.2m step 02850/05568 (51.19%) | loss: 2.553576 | lrm: 0.76 | dt: 1194.08ms | tok/sec: 878,145 | bf16_mfu: 53.94 | total time: 56.66m | eta: 54.2m step 02851/05568 (51.20%) | loss: 2.548031 | lrm: 0.76 | dt: 1191.87ms | tok/sec: 879,773 | bf16_mfu: 54.04 | total time: 56.68m | eta: 54.2m step 02852/05568 (51.22%) | loss: 2.549450 | lrm: 0.76 | dt: 1197.69ms | tok/sec: 875,499 | bf16_mfu: 53.78 | total time: 56.70m | eta: 54.2m step 02853/05568 (51.24%) | loss: 2.557021 | lrm: 0.76 | dt: 1194.24ms | tok/sec: 878,027 | bf16_mfu: 53.94 | total time: 56.72m | eta: 54.1m step 02854/05568 (51.26%) | loss: 2.558551 | lrm: 0.76 | dt: 1198.00ms | tok/sec: 875,272 | bf16_mfu: 53.77 | total time: 56.74m | eta: 54.1m step 02855/05568 (51.28%) | loss: 2.559510 | lrm: 0.76 | dt: 1191.74ms | tok/sec: 879,870 | bf16_mfu: 54.05 | total time: 56.76m | eta: 54.1m step 02856/05568 (51.29%) | loss: 2.550676 | lrm: 0.76 | dt: 1192.81ms | tok/sec: 879,080 | bf16_mfu: 54.00 | total time: 56.78m | eta: 54.1m step 02857/05568 (51.31%) | loss: 2.555090 | lrm: 0.76 | dt: 1199.00ms | tok/sec: 874,545 | bf16_mfu: 53.72 | total time: 56.80m | eta: 54.1m step 02858/05568 (51.33%) | loss: 2.548651 | lrm: 0.76 | dt: 1191.64ms | tok/sec: 879,939 | bf16_mfu: 54.05 | total time: 56.82m | eta: 54.0m step 02859/05568 (51.35%) | loss: 2.550106 | lrm: 0.76 | dt: 1189.79ms | tok/sec: 881,312 | bf16_mfu: 54.14 | total time: 56.84m | eta: 54.0m step 02860/05568 (51.36%) | loss: 2.564207 | lrm: 0.76 | dt: 1193.16ms | tok/sec: 878,822 | bf16_mfu: 53.98 | total time: 56.86m | eta: 54.0m step 02861/05568 (51.38%) | loss: 2.565904 | lrm: 0.76 | dt: 1193.71ms | tok/sec: 878,419 | bf16_mfu: 53.96 | total time: 56.88m | eta: 54.0m step 02862/05568 (51.40%) | loss: 2.560297 | lrm: 0.76 | dt: 1194.04ms | tok/sec: 878,176 | bf16_mfu: 53.94 | total time: 56.90m | eta: 54.0m step 02863/05568 (51.42%) | loss: 2.550369 | lrm: 0.76 | dt: 1196.49ms | tok/sec: 876,375 | bf16_mfu: 53.83 | total time: 56.92m | eta: 53.9m step 02864/05568 (51.44%) | loss: 2.545217 | lrm: 0.76 | dt: 1194.24ms | tok/sec: 878,025 | bf16_mfu: 53.94 | total time: 56.94m | eta: 53.9m step 02865/05568 (51.45%) | loss: 2.532586 | lrm: 0.76 | dt: 1194.70ms | tok/sec: 877,693 | bf16_mfu: 53.91 | total time: 56.96m | eta: 53.9m step 02866/05568 (51.47%) | loss: 2.525718 | lrm: 0.76 | dt: 1191.34ms | tok/sec: 880,165 | bf16_mfu: 54.07 | total time: 56.98m | eta: 53.9m step 02867/05568 (51.49%) | loss: 2.530398 | lrm: 0.76 | dt: 1193.11ms | tok/sec: 878,861 | bf16_mfu: 53.99 | total time: 57.00m | eta: 53.9m step 02868/05568 (51.51%) | loss: 2.529237 | lrm: 0.76 | dt: 1195.93ms | tok/sec: 876,786 | bf16_mfu: 53.86 | total time: 57.02m | eta: 53.8m step 02869/05568 (51.53%) | loss: 2.539650 | lrm: 0.76 | dt: 1194.70ms | tok/sec: 877,690 | bf16_mfu: 53.91 | total time: 57.04m | eta: 53.8m step 02870/05568 (51.54%) | loss: 2.547935 | lrm: 0.76 | dt: 1191.12ms | tok/sec: 880,324 | bf16_mfu: 54.08 | total time: 57.06m | eta: 53.8m step 02871/05568 (51.56%) | loss: 2.553764 | lrm: 0.76 | dt: 1197.88ms | tok/sec: 875,362 | bf16_mfu: 53.77 | total time: 57.08m | eta: 53.8m step 02872/05568 (51.58%) | loss: 2.561265 | lrm: 0.76 | dt: 1196.19ms | tok/sec: 876,595 | bf16_mfu: 53.85 | total time: 57.10m | eta: 53.8m step 02873/05568 (51.60%) | loss: 2.553765 | lrm: 0.76 | dt: 1198.01ms | tok/sec: 875,266 | bf16_mfu: 53.77 | total time: 57.12m | eta: 53.7m step 02874/05568 (51.62%) | loss: 2.564492 | lrm: 0.76 | dt: 1200.64ms | tok/sec: 873,348 | bf16_mfu: 53.65 | total time: 57.14m | eta: 53.7m step 02875/05568 (51.63%) | loss: 2.561068 | lrm: 0.76 | dt: 1196.98ms | tok/sec: 876,018 | bf16_mfu: 53.81 | total time: 57.16m | eta: 53.7m step 02876/05568 (51.65%) | loss: 2.556569 | lrm: 0.76 | dt: 1194.68ms | tok/sec: 877,706 | bf16_mfu: 53.92 | total time: 57.18m | eta: 53.7m step 02877/05568 (51.67%) | loss: 2.558551 | lrm: 0.76 | dt: 1190.34ms | tok/sec: 880,900 | bf16_mfu: 54.11 | total time: 57.20m | eta: 53.7m step 02878/05568 (51.69%) | loss: 2.558956 | lrm: 0.76 | dt: 1191.08ms | tok/sec: 880,358 | bf16_mfu: 54.08 | total time: 57.22m | eta: 53.6m step 02879/05568 (51.71%) | loss: 2.554039 | lrm: 0.76 | dt: 1189.82ms | tok/sec: 881,293 | bf16_mfu: 54.14 | total time: 57.24m | eta: 53.6m step 02880/05568 (51.72%) | loss: 2.554281 | lrm: 0.76 | dt: 1190.59ms | tok/sec: 880,721 | bf16_mfu: 54.10 | total time: 57.26m | eta: 53.6m step 02881/05568 (51.74%) | loss: 2.541239 | lrm: 0.76 | dt: 1193.45ms | tok/sec: 878,611 | bf16_mfu: 53.97 | total time: 57.28m | eta: 53.6m step 02882/05568 (51.76%) | loss: 2.550140 | lrm: 0.76 | dt: 1193.88ms | tok/sec: 878,293 | bf16_mfu: 53.95 | total time: 57.30m | eta: 53.6m step 02883/05568 (51.78%) | loss: 2.563576 | lrm: 0.75 | dt: 1196.46ms | tok/sec: 876,400 | bf16_mfu: 53.84 | total time: 57.32m | eta: 53.5m step 02884/05568 (51.80%) | loss: 2.555491 | lrm: 0.75 | dt: 1190.89ms | tok/sec: 880,494 | bf16_mfu: 54.09 | total time: 57.34m | eta: 53.5m step 02885/05568 (51.81%) | loss: 2.547589 | lrm: 0.75 | dt: 1194.47ms | tok/sec: 877,855 | bf16_mfu: 53.92 | total time: 57.36m | eta: 53.5m step 02886/05568 (51.83%) | loss: 2.550329 | lrm: 0.75 | dt: 1198.64ms | tok/sec: 874,805 | bf16_mfu: 53.74 | total time: 57.38m | eta: 53.5m step 02887/05568 (51.85%) | loss: 2.547364 | lrm: 0.75 | dt: 1193.33ms | tok/sec: 878,700 | bf16_mfu: 53.98 | total time: 57.40m | eta: 53.5m step 02888/05568 (51.87%) | loss: 2.538332 | lrm: 0.75 | dt: 1193.53ms | tok/sec: 878,550 | bf16_mfu: 53.97 | total time: 57.42m | eta: 53.4m step 02889/05568 (51.89%) | loss: 2.538075 | lrm: 0.75 | dt: 1196.15ms | tok/sec: 876,624 | bf16_mfu: 53.85 | total time: 57.44m | eta: 53.4m step 02890/05568 (51.90%) | loss: 2.542867 | lrm: 0.75 | dt: 1190.37ms | tok/sec: 880,879 | bf16_mfu: 54.11 | total time: 57.46m | eta: 53.4m step 02891/05568 (51.92%) | loss: 2.544525 | lrm: 0.75 | dt: 1191.21ms | tok/sec: 880,261 | bf16_mfu: 54.07 | total time: 57.48m | eta: 53.4m step 02892/05568 (51.94%) | loss: 2.552067 | lrm: 0.75 | dt: 1194.13ms | tok/sec: 878,107 | bf16_mfu: 53.94 | total time: 57.50m | eta: 53.4m step 02893/05568 (51.96%) | loss: 2.546359 | lrm: 0.75 | dt: 1189.87ms | tok/sec: 881,254 | bf16_mfu: 54.13 | total time: 57.52m | eta: 53.3m step 02894/05568 (51.98%) | loss: 2.546766 | lrm: 0.75 | dt: 1190.83ms | tok/sec: 880,539 | bf16_mfu: 54.09 | total time: 57.53m | eta: 53.3m step 02895/05568 (51.99%) | loss: 2.549230 | lrm: 0.75 | dt: 1194.75ms | tok/sec: 877,654 | bf16_mfu: 53.91 | total time: 57.55m | eta: 53.3m step 02896/05568 (52.01%) | loss: 2.546849 | lrm: 0.75 | dt: 1199.98ms | tok/sec: 873,826 | bf16_mfu: 53.68 | total time: 57.57m | eta: 53.3m step 02897/05568 (52.03%) | loss: 2.549820 | lrm: 0.75 | dt: 1193.36ms | tok/sec: 878,676 | bf16_mfu: 53.98 | total time: 57.59m | eta: 53.3m step 02898/05568 (52.05%) | loss: 2.546073 | lrm: 0.75 | dt: 1193.40ms | tok/sec: 878,648 | bf16_mfu: 53.97 | total time: 57.61m | eta: 53.2m step 02899/05568 (52.07%) | loss: 2.529685 | lrm: 0.75 | dt: 1192.84ms | tok/sec: 879,061 | bf16_mfu: 54.00 | total time: 57.63m | eta: 53.2m step 02900/05568 (52.08%) | loss: 2.534273 | lrm: 0.75 | dt: 1192.96ms | tok/sec: 878,969 | bf16_mfu: 53.99 | total time: 57.65m | eta: 53.2m step 02901/05568 (52.10%) | loss: 2.542642 | lrm: 0.75 | dt: 1190.48ms | tok/sec: 880,803 | bf16_mfu: 54.11 | total time: 57.67m | eta: 53.2m step 02902/05568 (52.12%) | loss: 2.528812 | lrm: 0.75 | dt: 1193.70ms | tok/sec: 878,423 | bf16_mfu: 53.96 | total time: 57.69m | eta: 53.2m step 02903/05568 (52.14%) | loss: 2.520093 | lrm: 0.75 | dt: 1190.82ms | tok/sec: 880,550 | bf16_mfu: 54.09 | total time: 57.71m | eta: 53.1m step 02904/05568 (52.16%) | loss: 2.525055 | lrm: 0.75 | dt: 1198.16ms | tok/sec: 875,158 | bf16_mfu: 53.76 | total time: 57.73m | eta: 53.1m step 02905/05568 (52.17%) | loss: 2.528462 | lrm: 0.75 | dt: 1191.51ms | tok/sec: 880,037 | bf16_mfu: 54.06 | total time: 57.75m | eta: 53.1m step 02906/05568 (52.19%) | loss: 2.536803 | lrm: 0.75 | dt: 1193.39ms | tok/sec: 878,654 | bf16_mfu: 53.97 | total time: 57.77m | eta: 53.1m step 02907/05568 (52.21%) | loss: 2.536793 | lrm: 0.75 | dt: 1199.36ms | tok/sec: 874,276 | bf16_mfu: 53.70 | total time: 57.79m | eta: 53.1m step 02908/05568 (52.23%) | loss: 2.532377 | lrm: 0.75 | dt: 1198.14ms | tok/sec: 875,171 | bf16_mfu: 53.76 | total time: 57.81m | eta: 53.0m step 02909/05568 (52.24%) | loss: 2.540476 | lrm: 0.75 | dt: 1195.65ms | tok/sec: 876,991 | bf16_mfu: 53.87 | total time: 57.83m | eta: 53.0m step 02910/05568 (52.26%) | loss: 2.545676 | lrm: 0.75 | dt: 1193.17ms | tok/sec: 878,815 | bf16_mfu: 53.98 | total time: 57.85m | eta: 53.0m step 02911/05568 (52.28%) | loss: 2.552609 | lrm: 0.75 | dt: 1194.82ms | tok/sec: 877,601 | bf16_mfu: 53.91 | total time: 57.87m | eta: 53.0m step 02912/05568 (52.30%) | loss: 2.543194 | lrm: 0.75 | dt: 1188.15ms | tok/sec: 882,530 | bf16_mfu: 54.21 | total time: 57.89m | eta: 53.0m step 02913/05568 (52.32%) | loss: 2.546535 | lrm: 0.75 | dt: 1194.76ms | tok/sec: 877,645 | bf16_mfu: 53.91 | total time: 57.91m | eta: 52.9m step 02914/05568 (52.33%) | loss: 2.545185 | lrm: 0.75 | dt: 1192.90ms | tok/sec: 879,014 | bf16_mfu: 54.00 | total time: 57.93m | eta: 52.9m step 02915/05568 (52.35%) | loss: 2.545601 | lrm: 0.75 | dt: 1198.01ms | tok/sec: 875,268 | bf16_mfu: 53.77 | total time: 57.95m | eta: 52.9m step 02916/05568 (52.37%) | loss: 2.560723 | lrm: 0.75 | dt: 1193.64ms | tok/sec: 878,472 | bf16_mfu: 53.96 | total time: 57.97m | eta: 52.9m step 02917/05568 (52.39%) | loss: 2.561554 | lrm: 0.75 | dt: 1195.55ms | tok/sec: 877,066 | bf16_mfu: 53.88 | total time: 57.99m | eta: 52.9m step 02918/05568 (52.41%) | loss: 2.546941 | lrm: 0.75 | dt: 1193.59ms | tok/sec: 878,509 | bf16_mfu: 53.96 | total time: 58.01m | eta: 52.8m step 02919/05568 (52.42%) | loss: 2.556377 | lrm: 0.75 | dt: 1196.83ms | tok/sec: 876,124 | bf16_mfu: 53.82 | total time: 58.03m | eta: 52.8m step 02920/05568 (52.44%) | loss: 2.558784 | lrm: 0.75 | dt: 1194.46ms | tok/sec: 877,867 | bf16_mfu: 53.93 | total time: 58.05m | eta: 52.8m step 02921/05568 (52.46%) | loss: 2.565019 | lrm: 0.74 | dt: 1191.24ms | tok/sec: 880,236 | bf16_mfu: 54.07 | total time: 58.07m | eta: 52.8m step 02922/05568 (52.48%) | loss: 2.558538 | lrm: 0.74 | dt: 1193.69ms | tok/sec: 878,433 | bf16_mfu: 53.96 | total time: 58.09m | eta: 52.8m step 02923/05568 (52.50%) | loss: 2.554877 | lrm: 0.74 | dt: 1193.13ms | tok/sec: 878,844 | bf16_mfu: 53.99 | total time: 58.11m | eta: 52.7m step 02924/05568 (52.51%) | loss: 2.550556 | lrm: 0.74 | dt: 1190.67ms | tok/sec: 880,661 | bf16_mfu: 54.10 | total time: 58.13m | eta: 52.7m step 02925/05568 (52.53%) | loss: 2.556969 | lrm: 0.74 | dt: 1196.26ms | tok/sec: 876,545 | bf16_mfu: 53.84 | total time: 58.15m | eta: 52.7m step 02926/05568 (52.55%) | loss: 2.557191 | lrm: 0.74 | dt: 1191.84ms | tok/sec: 879,795 | bf16_mfu: 54.04 | total time: 58.17m | eta: 52.7m step 02927/05568 (52.57%) | loss: 2.561288 | lrm: 0.74 | dt: 1197.84ms | tok/sec: 875,387 | bf16_mfu: 53.77 | total time: 58.19m | eta: 52.7m step 02928/05568 (52.59%) | loss: 2.552892 | lrm: 0.74 | dt: 1195.14ms | tok/sec: 877,365 | bf16_mfu: 53.89 | total time: 58.21m | eta: 52.6m step 02929/05568 (52.60%) | loss: 2.553002 | lrm: 0.74 | dt: 1192.94ms | tok/sec: 878,985 | bf16_mfu: 53.99 | total time: 58.23m | eta: 52.6m step 02930/05568 (52.62%) | loss: 2.542544 | lrm: 0.74 | dt: 1189.42ms | tok/sec: 881,582 | bf16_mfu: 54.15 | total time: 58.25m | eta: 52.6m step 02931/05568 (52.64%) | loss: 2.546872 | lrm: 0.74 | dt: 1190.51ms | tok/sec: 880,778 | bf16_mfu: 54.10 | total time: 58.27m | eta: 52.6m step 02932/05568 (52.66%) | loss: 2.545483 | lrm: 0.74 | dt: 1192.35ms | tok/sec: 879,420 | bf16_mfu: 54.02 | total time: 58.29m | eta: 52.6m step 02933/05568 (52.68%) | loss: 2.555450 | lrm: 0.74 | dt: 1197.27ms | tok/sec: 875,806 | bf16_mfu: 53.80 | total time: 58.31m | eta: 52.5m step 02934/05568 (52.69%) | loss: 2.549350 | lrm: 0.74 | dt: 1196.75ms | tok/sec: 876,183 | bf16_mfu: 53.82 | total time: 58.33m | eta: 52.5m step 02935/05568 (52.71%) | loss: 2.562004 | lrm: 0.74 | dt: 1191.30ms | tok/sec: 880,194 | bf16_mfu: 54.07 | total time: 58.35m | eta: 52.5m step 02936/05568 (52.73%) | loss: 2.561242 | lrm: 0.74 | dt: 1193.72ms | tok/sec: 878,410 | bf16_mfu: 53.96 | total time: 58.37m | eta: 52.5m step 02937/05568 (52.75%) | loss: 2.562082 | lrm: 0.74 | dt: 1191.27ms | tok/sec: 880,214 | bf16_mfu: 54.07 | total time: 58.39m | eta: 52.5m step 02938/05568 (52.77%) | loss: 2.548759 | lrm: 0.74 | dt: 1196.88ms | tok/sec: 876,088 | bf16_mfu: 53.82 | total time: 58.41m | eta: 52.4m step 02939/05568 (52.78%) | loss: 2.547048 | lrm: 0.74 | dt: 1188.39ms | tok/sec: 882,349 | bf16_mfu: 54.20 | total time: 58.43m | eta: 52.4m step 02940/05568 (52.80%) | loss: 2.554740 | lrm: 0.74 | dt: 1198.92ms | tok/sec: 874,599 | bf16_mfu: 53.72 | total time: 58.45m | eta: 52.4m step 02941/05568 (52.82%) | loss: 2.544041 | lrm: 0.74 | dt: 1197.79ms | tok/sec: 875,427 | bf16_mfu: 53.78 | total time: 58.47m | eta: 52.4m step 02942/05568 (52.84%) | loss: 2.547747 | lrm: 0.74 | dt: 1191.36ms | tok/sec: 880,152 | bf16_mfu: 54.07 | total time: 58.49m | eta: 52.4m step 02943/05568 (52.86%) | loss: 2.542045 | lrm: 0.74 | dt: 1194.88ms | tok/sec: 877,560 | bf16_mfu: 53.91 | total time: 58.51m | eta: 52.3m step 02944/05568 (52.87%) | loss: 2.546531 | lrm: 0.74 | dt: 1197.07ms | tok/sec: 875,954 | bf16_mfu: 53.81 | total time: 58.53m | eta: 52.3m step 02945/05568 (52.89%) | loss: 2.544891 | lrm: 0.74 | dt: 1192.84ms | tok/sec: 879,056 | bf16_mfu: 54.00 | total time: 58.55m | eta: 52.3m step 02946/05568 (52.91%) | loss: 2.546023 | lrm: 0.74 | dt: 1192.25ms | tok/sec: 879,495 | bf16_mfu: 54.03 | total time: 58.57m | eta: 52.3m step 02947/05568 (52.93%) | loss: 2.548686 | lrm: 0.74 | dt: 1193.49ms | tok/sec: 878,578 | bf16_mfu: 53.97 | total time: 58.59m | eta: 52.3m step 02948/05568 (52.95%) | loss: 2.553841 | lrm: 0.74 | dt: 1192.79ms | tok/sec: 879,091 | bf16_mfu: 54.00 | total time: 58.61m | eta: 52.2m step 02949/05568 (52.96%) | loss: 2.560749 | lrm: 0.74 | dt: 1193.72ms | tok/sec: 878,407 | bf16_mfu: 53.96 | total time: 58.63m | eta: 52.2m step 02950/05568 (52.98%) | loss: 2.552224 | lrm: 0.74 | dt: 1196.00ms | tok/sec: 876,733 | bf16_mfu: 53.86 | total time: 58.65m | eta: 52.2m step 02951/05568 (53.00%) | loss: 2.549992 | lrm: 0.74 | dt: 1192.84ms | tok/sec: 879,061 | bf16_mfu: 54.00 | total time: 58.67m | eta: 52.2m step 02952/05568 (53.02%) | loss: 2.557610 | lrm: 0.74 | dt: 1194.40ms | tok/sec: 877,909 | bf16_mfu: 53.93 | total time: 58.69m | eta: 52.2m step 02953/05568 (53.04%) | loss: 2.551975 | lrm: 0.74 | dt: 1190.98ms | tok/sec: 880,428 | bf16_mfu: 54.08 | total time: 58.71m | eta: 52.1m step 02954/05568 (53.05%) | loss: 2.547199 | lrm: 0.74 | dt: 1191.11ms | tok/sec: 880,334 | bf16_mfu: 54.08 | total time: 58.73m | eta: 52.1m step 02955/05568 (53.07%) | loss: 2.547447 | lrm: 0.74 | dt: 1192.17ms | tok/sec: 879,554 | bf16_mfu: 54.03 | total time: 58.75m | eta: 52.1m step 02956/05568 (53.09%) | loss: 2.550799 | lrm: 0.74 | dt: 1196.60ms | tok/sec: 876,296 | bf16_mfu: 53.83 | total time: 58.77m | eta: 52.1m step 02957/05568 (53.11%) | loss: 2.558015 | lrm: 0.74 | dt: 1196.69ms | tok/sec: 876,231 | bf16_mfu: 53.82 | total time: 58.79m | eta: 52.1m step 02958/05568 (53.12%) | loss: 2.557193 | lrm: 0.74 | dt: 1197.81ms | tok/sec: 875,412 | bf16_mfu: 53.77 | total time: 58.81m | eta: 52.0m step 02959/05568 (53.14%) | loss: 2.547993 | lrm: 0.73 | dt: 1191.69ms | tok/sec: 879,908 | bf16_mfu: 54.05 | total time: 58.83m | eta: 52.0m step 02960/05568 (53.16%) | loss: 2.541064 | lrm: 0.73 | dt: 1193.34ms | tok/sec: 878,691 | bf16_mfu: 53.98 | total time: 58.85m | eta: 52.0m step 02961/05568 (53.18%) | loss: 2.538459 | lrm: 0.73 | dt: 1194.89ms | tok/sec: 877,553 | bf16_mfu: 53.91 | total time: 58.87m | eta: 52.0m step 02962/05568 (53.20%) | loss: 2.551263 | lrm: 0.73 | dt: 1194.78ms | tok/sec: 877,631 | bf16_mfu: 53.91 | total time: 58.89m | eta: 52.0m step 02963/05568 (53.21%) | loss: 2.546677 | lrm: 0.73 | dt: 1195.76ms | tok/sec: 876,914 | bf16_mfu: 53.87 | total time: 58.91m | eta: 51.9m step 02964/05568 (53.23%) | loss: 2.548016 | lrm: 0.73 | dt: 1195.37ms | tok/sec: 877,200 | bf16_mfu: 53.88 | total time: 58.93m | eta: 51.9m step 02965/05568 (53.25%) | loss: 2.542850 | lrm: 0.73 | dt: 1196.93ms | tok/sec: 876,053 | bf16_mfu: 53.81 | total time: 58.95m | eta: 51.9m step 02966/05568 (53.27%) | loss: 2.545904 | lrm: 0.73 | dt: 1190.35ms | tok/sec: 880,893 | bf16_mfu: 54.11 | total time: 58.97m | eta: 51.9m step 02967/05568 (53.29%) | loss: 2.549590 | lrm: 0.73 | dt: 1192.04ms | tok/sec: 879,647 | bf16_mfu: 54.03 | total time: 58.99m | eta: 51.9m step 02968/05568 (53.30%) | loss: 2.555775 | lrm: 0.73 | dt: 1195.08ms | tok/sec: 877,408 | bf16_mfu: 53.90 | total time: 59.01m | eta: 51.8m step 02969/05568 (53.32%) | loss: 2.551284 | lrm: 0.73 | dt: 1187.95ms | tok/sec: 882,679 | bf16_mfu: 54.22 | total time: 59.03m | eta: 51.8m step 02970/05568 (53.34%) | loss: 2.552868 | lrm: 0.73 | dt: 1196.80ms | tok/sec: 876,151 | bf16_mfu: 53.82 | total time: 59.05m | eta: 51.8m step 02971/05568 (53.36%) | loss: 2.548130 | lrm: 0.73 | dt: 1193.59ms | tok/sec: 878,502 | bf16_mfu: 53.96 | total time: 59.07m | eta: 51.8m step 02972/05568 (53.38%) | loss: 2.561502 | lrm: 0.73 | dt: 1199.34ms | tok/sec: 874,292 | bf16_mfu: 53.71 | total time: 59.09m | eta: 51.8m step 02973/05568 (53.39%) | loss: 2.556666 | lrm: 0.73 | dt: 1195.43ms | tok/sec: 877,152 | bf16_mfu: 53.88 | total time: 59.11m | eta: 51.7m step 02974/05568 (53.41%) | loss: 2.550658 | lrm: 0.73 | dt: 1195.60ms | tok/sec: 877,032 | bf16_mfu: 53.87 | total time: 59.13m | eta: 51.7m step 02975/05568 (53.43%) | loss: 2.545448 | lrm: 0.73 | dt: 1194.09ms | tok/sec: 878,140 | bf16_mfu: 53.94 | total time: 59.15m | eta: 51.7m step 02976/05568 (53.45%) | loss: 2.542273 | lrm: 0.73 | dt: 1193.76ms | tok/sec: 878,378 | bf16_mfu: 53.96 | total time: 59.17m | eta: 51.7m step 02977/05568 (53.47%) | loss: 2.541781 | lrm: 0.73 | dt: 1197.25ms | tok/sec: 875,821 | bf16_mfu: 53.80 | total time: 59.19m | eta: 51.7m step 02978/05568 (53.48%) | loss: 2.545095 | lrm: 0.73 | dt: 1193.79ms | tok/sec: 878,356 | bf16_mfu: 53.96 | total time: 59.21m | eta: 51.6m step 02979/05568 (53.50%) | loss: 2.549257 | lrm: 0.73 | dt: 1196.23ms | tok/sec: 876,568 | bf16_mfu: 53.85 | total time: 59.23m | eta: 51.6m step 02980/05568 (53.52%) | loss: 2.549417 | lrm: 0.73 | dt: 1193.21ms | tok/sec: 878,783 | bf16_mfu: 53.98 | total time: 59.25m | eta: 51.6m step 02981/05568 (53.54%) | loss: 2.547170 | lrm: 0.73 | dt: 1193.17ms | tok/sec: 878,817 | bf16_mfu: 53.98 | total time: 59.27m | eta: 51.6m step 02982/05568 (53.56%) | loss: 2.540421 | lrm: 0.73 | dt: 1194.18ms | tok/sec: 878,070 | bf16_mfu: 53.94 | total time: 59.29m | eta: 51.6m step 02983/05568 (53.57%) | loss: 2.538916 | lrm: 0.73 | dt: 1192.91ms | tok/sec: 879,008 | bf16_mfu: 54.00 | total time: 59.31m | eta: 51.5m step 02984/05568 (53.59%) | loss: 2.556981 | lrm: 0.73 | dt: 1193.63ms | tok/sec: 878,474 | bf16_mfu: 53.96 | total time: 59.33m | eta: 51.5m step 02985/05568 (53.61%) | loss: 2.558985 | lrm: 0.73 | dt: 1190.58ms | tok/sec: 880,728 | bf16_mfu: 54.10 | total time: 59.35m | eta: 51.5m step 02986/05568 (53.63%) | loss: 2.547657 | lrm: 0.73 | dt: 1196.81ms | tok/sec: 876,140 | bf16_mfu: 53.82 | total time: 59.37m | eta: 51.5m step 02987/05568 (53.65%) | loss: 2.533662 | lrm: 0.73 | dt: 1194.67ms | tok/sec: 877,715 | bf16_mfu: 53.92 | total time: 59.39m | eta: 51.5m step 02988/05568 (53.66%) | loss: 2.521966 | lrm: 0.73 | dt: 1191.73ms | tok/sec: 879,875 | bf16_mfu: 54.05 | total time: 59.41m | eta: 51.4m step 02989/05568 (53.68%) | loss: 2.516057 | lrm: 0.73 | dt: 1196.46ms | tok/sec: 876,396 | bf16_mfu: 53.84 | total time: 59.43m | eta: 51.4m step 02990/05568 (53.70%) | loss: 2.509975 | lrm: 0.73 | dt: 1194.47ms | tok/sec: 877,858 | bf16_mfu: 53.92 | total time: 59.45m | eta: 51.4m step 02991/05568 (53.72%) | loss: 2.514521 | lrm: 0.73 | dt: 1191.07ms | tok/sec: 880,364 | bf16_mfu: 54.08 | total time: 59.47m | eta: 51.4m step 02992/05568 (53.74%) | loss: 2.526689 | lrm: 0.73 | dt: 1195.65ms | tok/sec: 876,990 | bf16_mfu: 53.87 | total time: 59.49m | eta: 51.4m step 02993/05568 (53.75%) | loss: 2.539873 | lrm: 0.73 | dt: 1197.13ms | tok/sec: 875,905 | bf16_mfu: 53.80 | total time: 59.51m | eta: 51.3m step 02994/05568 (53.77%) | loss: 2.545600 | lrm: 0.73 | dt: 1193.75ms | tok/sec: 878,390 | bf16_mfu: 53.96 | total time: 59.53m | eta: 51.3m step 02995/05568 (53.79%) | loss: 2.552142 | lrm: 0.73 | dt: 1189.63ms | tok/sec: 881,433 | bf16_mfu: 54.14 | total time: 59.54m | eta: 51.3m step 02996/05568 (53.81%) | loss: 2.547495 | lrm: 0.73 | dt: 1196.39ms | tok/sec: 876,450 | bf16_mfu: 53.84 | total time: 59.56m | eta: 51.3m step 02997/05568 (53.83%) | loss: 2.541742 | lrm: 0.72 | dt: 1192.19ms | tok/sec: 879,536 | bf16_mfu: 54.03 | total time: 59.58m | eta: 51.3m step 02998/05568 (53.84%) | loss: 2.543372 | lrm: 0.72 | dt: 1195.98ms | tok/sec: 876,750 | bf16_mfu: 53.86 | total time: 59.60m | eta: 51.2m step 02999/05568 (53.86%) | loss: 2.535946 | lrm: 0.72 | dt: 1193.62ms | tok/sec: 878,484 | bf16_mfu: 53.96 | total time: 59.62m | eta: 51.2m step:3000/5568 val_bpb:0.773139 val_time:3.88s step 03000/05568 (53.88%) | loss: 2.544506 | lrm: 0.72 | dt: 1195.01ms | tok/sec: 877,463 | bf16_mfu: 53.90 | total time: 59.64m | eta: 51.2m step 03001/05568 (53.90%) | loss: 2.544881 | lrm: 0.72 | dt: 1191.56ms | tok/sec: 880,000 | bf16_mfu: 54.06 | total time: 59.66m | eta: 51.2m step 03002/05568 (53.92%) | loss: 2.550384 | lrm: 0.72 | dt: 1191.98ms | tok/sec: 879,691 | bf16_mfu: 54.04 | total time: 59.68m | eta: 51.2m step 03003/05568 (53.93%) | loss: 2.547417 | lrm: 0.72 | dt: 1197.36ms | tok/sec: 875,742 | bf16_mfu: 53.79 | total time: 59.70m | eta: 51.1m step 03004/05568 (53.95%) | loss: 2.531706 | lrm: 0.72 | dt: 1189.30ms | tok/sec: 881,671 | bf16_mfu: 54.16 | total time: 59.72m | eta: 51.1m step 03005/05568 (53.97%) | loss: 2.537321 | lrm: 0.72 | dt: 1193.74ms | tok/sec: 878,398 | bf16_mfu: 53.96 | total time: 59.74m | eta: 51.1m step 03006/05568 (53.99%) | loss: 2.533787 | lrm: 0.72 | dt: 1192.99ms | tok/sec: 878,949 | bf16_mfu: 53.99 | total time: 59.76m | eta: 51.1m step 03007/05568 (54.01%) | loss: 2.540393 | lrm: 0.72 | dt: 1191.23ms | tok/sec: 880,242 | bf16_mfu: 54.07 | total time: 59.78m | eta: 51.1m step 03008/05568 (54.02%) | loss: 2.528139 | lrm: 0.72 | dt: 1191.80ms | tok/sec: 879,827 | bf16_mfu: 54.05 | total time: 59.80m | eta: 51.0m step 03009/05568 (54.04%) | loss: 2.544419 | lrm: 0.72 | dt: 1191.22ms | tok/sec: 880,251 | bf16_mfu: 54.07 | total time: 59.82m | eta: 51.0m step 03010/05568 (54.06%) | loss: 2.550516 | lrm: 0.72 | dt: 1192.64ms | tok/sec: 879,203 | bf16_mfu: 54.01 | total time: 59.84m | eta: 51.0m step 03011/05568 (54.08%) | loss: 2.559295 | lrm: 0.72 | dt: 1195.33ms | tok/sec: 877,225 | bf16_mfu: 53.89 | total time: 59.86m | eta: 51.0m step 03012/05568 (54.09%) | loss: 2.564689 | lrm: 0.72 | dt: 1185.02ms | tok/sec: 884,860 | bf16_mfu: 54.35 | total time: 59.88m | eta: 51.0m step 03013/05568 (54.11%) | loss: 2.572780 | lrm: 0.72 | dt: 1195.88ms | tok/sec: 876,820 | bf16_mfu: 53.86 | total time: 59.90m | eta: 50.9m step 03014/05568 (54.13%) | loss: 2.565964 | lrm: 0.72 | dt: 1189.06ms | tok/sec: 881,853 | bf16_mfu: 54.17 | total time: 59.92m | eta: 50.9m step 03015/05568 (54.15%) | loss: 2.556919 | lrm: 0.72 | dt: 1194.87ms | tok/sec: 877,562 | bf16_mfu: 53.91 | total time: 59.94m | eta: 50.9m step 03016/05568 (54.17%) | loss: 2.549434 | lrm: 0.72 | dt: 1191.82ms | tok/sec: 879,808 | bf16_mfu: 54.04 | total time: 59.96m | eta: 50.9m step 03017/05568 (54.18%) | loss: 2.555544 | lrm: 0.72 | dt: 1190.25ms | tok/sec: 880,970 | bf16_mfu: 54.12 | total time: 59.98m | eta: 50.9m step 03018/05568 (54.20%) | loss: 2.550556 | lrm: 0.72 | dt: 1194.59ms | tok/sec: 877,771 | bf16_mfu: 53.92 | total time: 60.00m | eta: 50.8m step 03019/05568 (54.22%) | loss: 2.548367 | lrm: 0.72 | dt: 1187.33ms | tok/sec: 883,137 | bf16_mfu: 54.25 | total time: 60.02m | eta: 50.8m step 03020/05568 (54.24%) | loss: 2.538939 | lrm: 0.72 | dt: 1190.79ms | tok/sec: 880,570 | bf16_mfu: 54.09 | total time: 60.04m | eta: 50.8m step 03021/05568 (54.26%) | loss: 2.534120 | lrm: 0.72 | dt: 1190.81ms | tok/sec: 880,559 | bf16_mfu: 54.09 | total time: 60.06m | eta: 50.8m step 03022/05568 (54.27%) | loss: 2.523404 | lrm: 0.72 | dt: 1193.07ms | tok/sec: 878,888 | bf16_mfu: 53.99 | total time: 60.08m | eta: 50.8m step 03023/05568 (54.29%) | loss: 2.523671 | lrm: 0.72 | dt: 1192.35ms | tok/sec: 879,422 | bf16_mfu: 54.02 | total time: 60.10m | eta: 50.7m step 03024/05568 (54.31%) | loss: 2.514651 | lrm: 0.72 | dt: 1194.11ms | tok/sec: 878,120 | bf16_mfu: 53.94 | total time: 60.12m | eta: 50.7m step 03025/05568 (54.33%) | loss: 2.524411 | lrm: 0.72 | dt: 1191.90ms | tok/sec: 879,752 | bf16_mfu: 54.04 | total time: 60.14m | eta: 50.7m step 03026/05568 (54.35%) | loss: 2.511223 | lrm: 0.72 | dt: 1195.09ms | tok/sec: 877,404 | bf16_mfu: 53.90 | total time: 60.16m | eta: 50.7m step 03027/05568 (54.36%) | loss: 2.513332 | lrm: 0.72 | dt: 1189.05ms | tok/sec: 881,859 | bf16_mfu: 54.17 | total time: 60.18m | eta: 50.7m step 03028/05568 (54.38%) | loss: 2.517236 | lrm: 0.72 | dt: 1192.64ms | tok/sec: 879,204 | bf16_mfu: 54.01 | total time: 60.20m | eta: 50.6m step 03029/05568 (54.40%) | loss: 2.527619 | lrm: 0.72 | dt: 1193.90ms | tok/sec: 878,274 | bf16_mfu: 53.95 | total time: 60.22m | eta: 50.6m step 03030/05568 (54.42%) | loss: 2.522386 | lrm: 0.72 | dt: 1194.31ms | tok/sec: 877,977 | bf16_mfu: 53.93 | total time: 60.24m | eta: 50.6m step 03031/05568 (54.44%) | loss: 2.519135 | lrm: 0.72 | dt: 1189.08ms | tok/sec: 881,839 | bf16_mfu: 54.17 | total time: 60.26m | eta: 50.6m step 03032/05568 (54.45%) | loss: 2.526044 | lrm: 0.72 | dt: 1191.86ms | tok/sec: 879,779 | bf16_mfu: 54.04 | total time: 60.28m | eta: 50.6m step 03033/05568 (54.47%) | loss: 2.517877 | lrm: 0.72 | dt: 1198.45ms | tok/sec: 874,944 | bf16_mfu: 53.75 | total time: 60.30m | eta: 50.5m step 03034/05568 (54.49%) | loss: 2.515589 | lrm: 0.72 | dt: 1193.35ms | tok/sec: 878,679 | bf16_mfu: 53.98 | total time: 60.32m | eta: 50.5m step 03035/05568 (54.51%) | loss: 2.516193 | lrm: 0.71 | dt: 1196.18ms | tok/sec: 876,603 | bf16_mfu: 53.85 | total time: 60.34m | eta: 50.5m step 03036/05568 (54.53%) | loss: 2.501736 | lrm: 0.71 | dt: 1197.38ms | tok/sec: 875,726 | bf16_mfu: 53.79 | total time: 60.36m | eta: 50.5m step 03037/05568 (54.54%) | loss: 2.502170 | lrm: 0.71 | dt: 1192.54ms | tok/sec: 879,280 | bf16_mfu: 54.01 | total time: 60.38m | eta: 50.5m step 03038/05568 (54.56%) | loss: 2.499198 | lrm: 0.71 | dt: 1194.54ms | tok/sec: 877,804 | bf16_mfu: 53.92 | total time: 60.40m | eta: 50.4m step 03039/05568 (54.58%) | loss: 2.503251 | lrm: 0.71 | dt: 1194.35ms | tok/sec: 877,947 | bf16_mfu: 53.93 | total time: 60.42m | eta: 50.4m step 03040/05568 (54.60%) | loss: 2.494160 | lrm: 0.71 | dt: 1195.03ms | tok/sec: 877,448 | bf16_mfu: 53.90 | total time: 60.44m | eta: 50.4m step 03041/05568 (54.62%) | loss: 2.490150 | lrm: 0.71 | dt: 1190.72ms | tok/sec: 880,621 | bf16_mfu: 54.09 | total time: 60.46m | eta: 50.4m step 03042/05568 (54.63%) | loss: 2.492011 | lrm: 0.71 | dt: 1195.99ms | tok/sec: 876,740 | bf16_mfu: 53.86 | total time: 60.48m | eta: 50.4m step 03043/05568 (54.65%) | loss: 2.485756 | lrm: 0.71 | dt: 1197.31ms | tok/sec: 875,775 | bf16_mfu: 53.80 | total time: 60.50m | eta: 50.3m step 03044/05568 (54.67%) | loss: 2.499297 | lrm: 0.71 | dt: 1189.53ms | tok/sec: 881,501 | bf16_mfu: 54.15 | total time: 60.52m | eta: 50.3m step 03045/05568 (54.69%) | loss: 2.508339 | lrm: 0.71 | dt: 1195.77ms | tok/sec: 876,905 | bf16_mfu: 53.87 | total time: 60.54m | eta: 50.3m step 03046/05568 (54.71%) | loss: 2.507080 | lrm: 0.71 | dt: 1191.82ms | tok/sec: 879,814 | bf16_mfu: 54.04 | total time: 60.56m | eta: 50.3m step 03047/05568 (54.72%) | loss: 2.506075 | lrm: 0.71 | dt: 1194.59ms | tok/sec: 877,772 | bf16_mfu: 53.92 | total time: 60.58m | eta: 50.3m step 03048/05568 (54.74%) | loss: 2.507203 | lrm: 0.71 | dt: 1191.50ms | tok/sec: 880,043 | bf16_mfu: 54.06 | total time: 60.60m | eta: 50.2m step 03049/05568 (54.76%) | loss: 2.522641 | lrm: 0.71 | dt: 1195.32ms | tok/sec: 877,236 | bf16_mfu: 53.89 | total time: 60.62m | eta: 50.2m step 03050/05568 (54.78%) | loss: 2.526221 | lrm: 0.71 | dt: 1191.52ms | tok/sec: 880,028 | bf16_mfu: 54.06 | total time: 60.64m | eta: 50.2m step 03051/05568 (54.80%) | loss: 2.520036 | lrm: 0.71 | dt: 1192.80ms | tok/sec: 879,085 | bf16_mfu: 54.00 | total time: 60.66m | eta: 50.2m step 03052/05568 (54.81%) | loss: 2.513832 | lrm: 0.71 | dt: 1191.90ms | tok/sec: 879,750 | bf16_mfu: 54.04 | total time: 60.68m | eta: 50.2m step 03053/05568 (54.83%) | loss: 2.517988 | lrm: 0.71 | dt: 1189.04ms | tok/sec: 881,866 | bf16_mfu: 54.17 | total time: 60.70m | eta: 50.1m step 03054/05568 (54.85%) | loss: 2.517056 | lrm: 0.71 | dt: 1192.24ms | tok/sec: 879,502 | bf16_mfu: 54.03 | total time: 60.72m | eta: 50.1m step 03055/05568 (54.87%) | loss: 2.528800 | lrm: 0.71 | dt: 1195.03ms | tok/sec: 877,444 | bf16_mfu: 53.90 | total time: 60.74m | eta: 50.1m step 03056/05568 (54.89%) | loss: 2.523383 | lrm: 0.71 | dt: 1192.17ms | tok/sec: 879,555 | bf16_mfu: 54.03 | total time: 60.76m | eta: 50.1m step 03057/05568 (54.90%) | loss: 2.514627 | lrm: 0.71 | dt: 1194.89ms | tok/sec: 877,547 | bf16_mfu: 53.91 | total time: 60.78m | eta: 50.1m step 03058/05568 (54.92%) | loss: 2.511043 | lrm: 0.71 | dt: 1198.57ms | tok/sec: 874,856 | bf16_mfu: 53.74 | total time: 60.80m | eta: 50.0m step 03059/05568 (54.94%) | loss: 2.510687 | lrm: 0.71 | dt: 1188.34ms | tok/sec: 882,384 | bf16_mfu: 54.20 | total time: 60.82m | eta: 50.0m step 03060/05568 (54.96%) | loss: 2.512047 | lrm: 0.71 | dt: 1192.52ms | tok/sec: 879,292 | bf16_mfu: 54.01 | total time: 60.84m | eta: 50.0m step 03061/05568 (54.97%) | loss: 2.509210 | lrm: 0.71 | dt: 1191.66ms | tok/sec: 879,930 | bf16_mfu: 54.05 | total time: 60.86m | eta: 50.0m step 03062/05568 (54.99%) | loss: 2.513685 | lrm: 0.71 | dt: 1190.88ms | tok/sec: 880,505 | bf16_mfu: 54.09 | total time: 60.88m | eta: 50.0m step 03063/05568 (55.01%) | loss: 2.528527 | lrm: 0.71 | dt: 1190.01ms | tok/sec: 881,148 | bf16_mfu: 54.13 | total time: 60.90m | eta: 49.9m step 03064/05568 (55.03%) | loss: 2.542717 | lrm: 0.71 | dt: 1195.56ms | tok/sec: 877,055 | bf16_mfu: 53.88 | total time: 60.92m | eta: 49.9m step 03065/05568 (55.05%) | loss: 2.547677 | lrm: 0.71 | dt: 1187.83ms | tok/sec: 882,763 | bf16_mfu: 54.23 | total time: 60.94m | eta: 49.9m step 03066/05568 (55.06%) | loss: 2.539595 | lrm: 0.71 | dt: 1196.80ms | tok/sec: 876,151 | bf16_mfu: 53.82 | total time: 60.96m | eta: 49.9m step 03067/05568 (55.08%) | loss: 2.540726 | lrm: 0.71 | dt: 1186.37ms | tok/sec: 883,854 | bf16_mfu: 54.29 | total time: 60.98m | eta: 49.9m step 03068/05568 (55.10%) | loss: 2.544845 | lrm: 0.71 | dt: 1192.09ms | tok/sec: 879,613 | bf16_mfu: 54.03 | total time: 61.00m | eta: 49.8m step 03069/05568 (55.12%) | loss: 2.536243 | lrm: 0.71 | dt: 1192.92ms | tok/sec: 879,000 | bf16_mfu: 53.99 | total time: 61.02m | eta: 49.8m step 03070/05568 (55.14%) | loss: 2.541118 | lrm: 0.71 | dt: 1189.94ms | tok/sec: 881,203 | bf16_mfu: 54.13 | total time: 61.04m | eta: 49.8m step 03071/05568 (55.15%) | loss: 2.534428 | lrm: 0.71 | dt: 1188.59ms | tok/sec: 882,201 | bf16_mfu: 54.19 | total time: 61.06m | eta: 49.8m step 03072/05568 (55.17%) | loss: 2.522375 | lrm: 0.71 | dt: 1192.42ms | tok/sec: 879,365 | bf16_mfu: 54.02 | total time: 61.08m | eta: 49.8m step 03073/05568 (55.19%) | loss: 2.512886 | lrm: 0.70 | dt: 1192.93ms | tok/sec: 878,993 | bf16_mfu: 53.99 | total time: 61.10m | eta: 49.7m step 03074/05568 (55.21%) | loss: 2.513767 | lrm: 0.70 | dt: 1190.85ms | tok/sec: 880,526 | bf16_mfu: 54.09 | total time: 61.12m | eta: 49.7m step 03075/05568 (55.23%) | loss: 2.526132 | lrm: 0.70 | dt: 1191.19ms | tok/sec: 880,272 | bf16_mfu: 54.07 | total time: 61.14m | eta: 49.7m step 03076/05568 (55.24%) | loss: 2.532518 | lrm: 0.70 | dt: 1188.67ms | tok/sec: 882,145 | bf16_mfu: 54.19 | total time: 61.15m | eta: 49.7m step 03077/05568 (55.26%) | loss: 2.527025 | lrm: 0.70 | dt: 1191.22ms | tok/sec: 880,252 | bf16_mfu: 54.07 | total time: 61.17m | eta: 49.7m step 03078/05568 (55.28%) | loss: 2.530460 | lrm: 0.70 | dt: 1194.31ms | tok/sec: 877,976 | bf16_mfu: 53.93 | total time: 61.19m | eta: 49.6m step 03079/05568 (55.30%) | loss: 2.530339 | lrm: 0.70 | dt: 1191.51ms | tok/sec: 880,039 | bf16_mfu: 54.06 | total time: 61.21m | eta: 49.6m step 03080/05568 (55.32%) | loss: 2.522786 | lrm: 0.70 | dt: 1189.69ms | tok/sec: 881,387 | bf16_mfu: 54.14 | total time: 61.23m | eta: 49.6m step 03081/05568 (55.33%) | loss: 2.510991 | lrm: 0.70 | dt: 1190.58ms | tok/sec: 880,728 | bf16_mfu: 54.10 | total time: 61.25m | eta: 49.6m step 03082/05568 (55.35%) | loss: 2.524335 | lrm: 0.70 | dt: 1188.72ms | tok/sec: 882,107 | bf16_mfu: 54.19 | total time: 61.27m | eta: 49.6m step 03083/05568 (55.37%) | loss: 2.525563 | lrm: 0.70 | dt: 1197.65ms | tok/sec: 875,530 | bf16_mfu: 53.78 | total time: 61.29m | eta: 49.5m step 03084/05568 (55.39%) | loss: 2.524546 | lrm: 0.70 | dt: 1197.43ms | tok/sec: 875,690 | bf16_mfu: 53.79 | total time: 61.31m | eta: 49.5m step 03085/05568 (55.41%) | loss: 2.526145 | lrm: 0.70 | dt: 1189.03ms | tok/sec: 881,872 | bf16_mfu: 54.17 | total time: 61.33m | eta: 49.5m step 03086/05568 (55.42%) | loss: 2.533260 | lrm: 0.70 | dt: 1197.28ms | tok/sec: 875,798 | bf16_mfu: 53.80 | total time: 61.35m | eta: 49.5m step 03087/05568 (55.44%) | loss: 2.545301 | lrm: 0.70 | dt: 1193.83ms | tok/sec: 878,327 | bf16_mfu: 53.95 | total time: 61.37m | eta: 49.5m step 03088/05568 (55.46%) | loss: 2.537201 | lrm: 0.70 | dt: 1193.11ms | tok/sec: 878,858 | bf16_mfu: 53.99 | total time: 61.39m | eta: 49.4m step 03089/05568 (55.48%) | loss: 2.542958 | lrm: 0.70 | dt: 1189.59ms | tok/sec: 881,462 | bf16_mfu: 54.15 | total time: 61.41m | eta: 49.4m step 03090/05568 (55.50%) | loss: 2.529142 | lrm: 0.70 | dt: 1191.60ms | tok/sec: 879,973 | bf16_mfu: 54.05 | total time: 61.43m | eta: 49.4m step 03091/05568 (55.51%) | loss: 2.531579 | lrm: 0.70 | dt: 1186.78ms | tok/sec: 883,548 | bf16_mfu: 54.27 | total time: 61.45m | eta: 49.4m step 03092/05568 (55.53%) | loss: 2.536857 | lrm: 0.70 | dt: 1189.21ms | tok/sec: 881,738 | bf16_mfu: 54.16 | total time: 61.47m | eta: 49.4m step 03093/05568 (55.55%) | loss: 2.526069 | lrm: 0.70 | dt: 1194.30ms | tok/sec: 877,985 | bf16_mfu: 53.93 | total time: 61.49m | eta: 49.3m step 03094/05568 (55.57%) | loss: 2.528564 | lrm: 0.70 | dt: 1191.36ms | tok/sec: 880,150 | bf16_mfu: 54.07 | total time: 61.51m | eta: 49.3m step 03095/05568 (55.59%) | loss: 2.528429 | lrm: 0.70 | dt: 1188.78ms | tok/sec: 882,063 | bf16_mfu: 54.18 | total time: 61.53m | eta: 49.3m step 03096/05568 (55.60%) | loss: 2.540597 | lrm: 0.70 | dt: 1191.09ms | tok/sec: 880,353 | bf16_mfu: 54.08 | total time: 61.55m | eta: 49.3m step 03097/05568 (55.62%) | loss: 2.538841 | lrm: 0.70 | dt: 1198.80ms | tok/sec: 874,687 | bf16_mfu: 53.73 | total time: 61.57m | eta: 49.3m step 03098/05568 (55.64%) | loss: 2.531160 | lrm: 0.70 | dt: 1189.87ms | tok/sec: 881,248 | bf16_mfu: 54.13 | total time: 61.59m | eta: 49.2m step 03099/05568 (55.66%) | loss: 2.533503 | lrm: 0.70 | dt: 1191.78ms | tok/sec: 879,838 | bf16_mfu: 54.05 | total time: 61.61m | eta: 49.2m step 03100/05568 (55.68%) | loss: 2.533692 | lrm: 0.70 | dt: 1188.11ms | tok/sec: 882,560 | bf16_mfu: 54.21 | total time: 61.63m | eta: 49.2m step 03101/05568 (55.69%) | loss: 2.536761 | lrm: 0.70 | dt: 1193.42ms | tok/sec: 878,629 | bf16_mfu: 53.97 | total time: 61.65m | eta: 49.2m step 03102/05568 (55.71%) | loss: 2.542804 | lrm: 0.70 | dt: 1191.49ms | tok/sec: 880,054 | bf16_mfu: 54.06 | total time: 61.67m | eta: 49.2m step 03103/05568 (55.73%) | loss: 2.558857 | lrm: 0.70 | dt: 1191.10ms | tok/sec: 880,346 | bf16_mfu: 54.08 | total time: 61.69m | eta: 49.1m step 03104/05568 (55.75%) | loss: 2.558649 | lrm: 0.70 | dt: 1194.67ms | tok/sec: 877,714 | bf16_mfu: 53.92 | total time: 61.71m | eta: 49.1m step 03105/05568 (55.77%) | loss: 2.547143 | lrm: 0.70 | dt: 1191.16ms | tok/sec: 880,295 | bf16_mfu: 54.07 | total time: 61.73m | eta: 49.1m step 03106/05568 (55.78%) | loss: 2.536473 | lrm: 0.70 | dt: 1191.42ms | tok/sec: 880,103 | bf16_mfu: 54.06 | total time: 61.75m | eta: 49.1m step 03107/05568 (55.80%) | loss: 2.541916 | lrm: 0.70 | dt: 1192.62ms | tok/sec: 879,223 | bf16_mfu: 54.01 | total time: 61.77m | eta: 49.1m step 03108/05568 (55.82%) | loss: 2.536909 | lrm: 0.70 | dt: 1192.29ms | tok/sec: 879,466 | bf16_mfu: 54.02 | total time: 61.79m | eta: 49.0m step 03109/05568 (55.84%) | loss: 2.526933 | lrm: 0.70 | dt: 1191.27ms | tok/sec: 880,213 | bf16_mfu: 54.07 | total time: 61.81m | eta: 49.0m step 03110/05568 (55.85%) | loss: 2.534388 | lrm: 0.70 | dt: 1194.84ms | tok/sec: 877,587 | bf16_mfu: 53.91 | total time: 61.83m | eta: 49.0m step 03111/05568 (55.87%) | loss: 2.533477 | lrm: 0.69 | dt: 1188.69ms | tok/sec: 882,126 | bf16_mfu: 54.19 | total time: 61.85m | eta: 49.0m step 03112/05568 (55.89%) | loss: 2.532076 | lrm: 0.69 | dt: 1189.62ms | tok/sec: 881,435 | bf16_mfu: 54.14 | total time: 61.87m | eta: 49.0m step 03113/05568 (55.91%) | loss: 2.533460 | lrm: 0.69 | dt: 1189.48ms | tok/sec: 881,540 | bf16_mfu: 54.15 | total time: 61.89m | eta: 48.9m step 03114/05568 (55.93%) | loss: 2.521375 | lrm: 0.69 | dt: 1196.12ms | tok/sec: 876,647 | bf16_mfu: 53.85 | total time: 61.91m | eta: 48.9m step 03115/05568 (55.94%) | loss: 2.519928 | lrm: 0.69 | dt: 1192.86ms | tok/sec: 879,046 | bf16_mfu: 54.00 | total time: 61.93m | eta: 48.9m step 03116/05568 (55.96%) | loss: 2.527376 | lrm: 0.69 | dt: 1190.15ms | tok/sec: 881,042 | bf16_mfu: 54.12 | total time: 61.95m | eta: 48.9m step 03117/05568 (55.98%) | loss: 2.534200 | lrm: 0.69 | dt: 1191.63ms | tok/sec: 879,951 | bf16_mfu: 54.05 | total time: 61.97m | eta: 48.9m step 03118/05568 (56.00%) | loss: 2.544951 | lrm: 0.69 | dt: 1194.23ms | tok/sec: 878,035 | bf16_mfu: 53.94 | total time: 61.99m | eta: 48.8m step 03119/05568 (56.02%) | loss: 2.531659 | lrm: 0.69 | dt: 1191.05ms | tok/sec: 880,378 | bf16_mfu: 54.08 | total time: 62.01m | eta: 48.8m step 03120/05568 (56.03%) | loss: 2.532414 | lrm: 0.69 | dt: 1194.72ms | tok/sec: 877,678 | bf16_mfu: 53.91 | total time: 62.03m | eta: 48.8m step 03121/05568 (56.05%) | loss: 2.536657 | lrm: 0.69 | dt: 1195.60ms | tok/sec: 877,029 | bf16_mfu: 53.87 | total time: 62.05m | eta: 48.8m step 03122/05568 (56.07%) | loss: 2.545230 | lrm: 0.69 | dt: 1191.08ms | tok/sec: 880,355 | bf16_mfu: 54.08 | total time: 62.07m | eta: 48.8m step 03123/05568 (56.09%) | loss: 2.542186 | lrm: 0.69 | dt: 1192.42ms | tok/sec: 879,364 | bf16_mfu: 54.02 | total time: 62.09m | eta: 48.7m step 03124/05568 (56.11%) | loss: 2.550765 | lrm: 0.69 | dt: 1187.48ms | tok/sec: 883,025 | bf16_mfu: 54.24 | total time: 62.11m | eta: 48.7m step 03125/05568 (56.12%) | loss: 2.554164 | lrm: 0.69 | dt: 1193.45ms | tok/sec: 878,608 | bf16_mfu: 53.97 | total time: 62.13m | eta: 48.7m step 03126/05568 (56.14%) | loss: 2.557551 | lrm: 0.69 | dt: 1193.26ms | tok/sec: 878,745 | bf16_mfu: 53.98 | total time: 62.15m | eta: 48.7m step 03127/05568 (56.16%) | loss: 2.555107 | lrm: 0.69 | dt: 1192.92ms | tok/sec: 878,996 | bf16_mfu: 53.99 | total time: 62.17m | eta: 48.7m step 03128/05568 (56.18%) | loss: 2.556060 | lrm: 0.69 | dt: 1190.74ms | tok/sec: 880,609 | bf16_mfu: 54.09 | total time: 62.19m | eta: 48.6m step 03129/05568 (56.20%) | loss: 2.544930 | lrm: 0.69 | dt: 1190.79ms | tok/sec: 880,574 | bf16_mfu: 54.09 | total time: 62.21m | eta: 48.6m step 03130/05568 (56.21%) | loss: 2.548012 | lrm: 0.69 | dt: 1192.88ms | tok/sec: 879,028 | bf16_mfu: 54.00 | total time: 62.23m | eta: 48.6m step 03131/05568 (56.23%) | loss: 2.558830 | lrm: 0.69 | dt: 1190.25ms | tok/sec: 880,973 | bf16_mfu: 54.12 | total time: 62.25m | eta: 48.6m step 03132/05568 (56.25%) | loss: 2.549657 | lrm: 0.69 | dt: 1197.55ms | tok/sec: 875,602 | bf16_mfu: 53.79 | total time: 62.27m | eta: 48.6m step 03133/05568 (56.27%) | loss: 2.543450 | lrm: 0.69 | dt: 1192.36ms | tok/sec: 879,414 | bf16_mfu: 54.02 | total time: 62.29m | eta: 48.5m step 03134/05568 (56.29%) | loss: 2.533015 | lrm: 0.69 | dt: 1197.48ms | tok/sec: 875,650 | bf16_mfu: 53.79 | total time: 62.31m | eta: 48.5m step 03135/05568 (56.30%) | loss: 2.535752 | lrm: 0.69 | dt: 1188.01ms | tok/sec: 882,634 | bf16_mfu: 54.22 | total time: 62.33m | eta: 48.5m step 03136/05568 (56.32%) | loss: 2.529596 | lrm: 0.69 | dt: 1194.93ms | tok/sec: 877,519 | bf16_mfu: 53.90 | total time: 62.35m | eta: 48.5m step 03137/05568 (56.34%) | loss: 2.527920 | lrm: 0.69 | dt: 1191.25ms | tok/sec: 880,230 | bf16_mfu: 54.07 | total time: 62.37m | eta: 48.5m step 03138/05568 (56.36%) | loss: 2.519011 | lrm: 0.69 | dt: 1190.51ms | tok/sec: 880,781 | bf16_mfu: 54.10 | total time: 62.39m | eta: 48.4m step 03139/05568 (56.38%) | loss: 2.526915 | lrm: 0.69 | dt: 1200.29ms | tok/sec: 873,600 | bf16_mfu: 53.66 | total time: 62.41m | eta: 48.4m step 03140/05568 (56.39%) | loss: 2.519899 | lrm: 0.69 | dt: 1190.07ms | tok/sec: 881,105 | bf16_mfu: 54.12 | total time: 62.43m | eta: 48.4m step 03141/05568 (56.41%) | loss: 2.522486 | lrm: 0.69 | dt: 1195.88ms | tok/sec: 876,824 | bf16_mfu: 53.86 | total time: 62.45m | eta: 48.4m step 03142/05568 (56.43%) | loss: 2.530436 | lrm: 0.69 | dt: 1192.76ms | tok/sec: 879,120 | bf16_mfu: 54.00 | total time: 62.47m | eta: 48.4m step 03143/05568 (56.45%) | loss: 2.531151 | lrm: 0.69 | dt: 1189.35ms | tok/sec: 881,636 | bf16_mfu: 54.16 | total time: 62.49m | eta: 48.3m step 03144/05568 (56.47%) | loss: 2.522725 | lrm: 0.69 | dt: 1191.84ms | tok/sec: 879,799 | bf16_mfu: 54.04 | total time: 62.51m | eta: 48.3m step 03145/05568 (56.48%) | loss: 2.528322 | lrm: 0.69 | dt: 1191.83ms | tok/sec: 879,800 | bf16_mfu: 54.04 | total time: 62.53m | eta: 48.3m step 03146/05568 (56.50%) | loss: 2.528163 | lrm: 0.69 | dt: 1194.81ms | tok/sec: 877,608 | bf16_mfu: 53.91 | total time: 62.55m | eta: 48.3m step 03147/05568 (56.52%) | loss: 2.525318 | lrm: 0.69 | dt: 1194.22ms | tok/sec: 878,045 | bf16_mfu: 53.94 | total time: 62.57m | eta: 48.3m step 03148/05568 (56.54%) | loss: 2.526240 | lrm: 0.69 | dt: 1193.27ms | tok/sec: 878,745 | bf16_mfu: 53.98 | total time: 62.59m | eta: 48.2m step 03149/05568 (56.56%) | loss: 2.523360 | lrm: 0.68 | dt: 1192.43ms | tok/sec: 879,357 | bf16_mfu: 54.02 | total time: 62.61m | eta: 48.2m step 03150/05568 (56.57%) | loss: 2.525984 | lrm: 0.68 | dt: 1191.47ms | tok/sec: 880,072 | bf16_mfu: 54.06 | total time: 62.63m | eta: 48.2m step 03151/05568 (56.59%) | loss: 2.521359 | lrm: 0.68 | dt: 1192.06ms | tok/sec: 879,636 | bf16_mfu: 54.03 | total time: 62.65m | eta: 48.2m step 03152/05568 (56.61%) | loss: 2.527033 | lrm: 0.68 | dt: 1189.23ms | tok/sec: 881,723 | bf16_mfu: 54.16 | total time: 62.67m | eta: 48.2m step 03153/05568 (56.63%) | loss: 2.526700 | lrm: 0.68 | dt: 1193.29ms | tok/sec: 878,723 | bf16_mfu: 53.98 | total time: 62.69m | eta: 48.1m step 03154/05568 (56.65%) | loss: 2.517043 | lrm: 0.68 | dt: 1190.59ms | tok/sec: 880,716 | bf16_mfu: 54.10 | total time: 62.70m | eta: 48.1m step 03155/05568 (56.66%) | loss: 2.507701 | lrm: 0.68 | dt: 1193.16ms | tok/sec: 878,821 | bf16_mfu: 53.98 | total time: 62.72m | eta: 48.1m step 03156/05568 (56.68%) | loss: 2.509597 | lrm: 0.68 | dt: 1194.72ms | tok/sec: 877,678 | bf16_mfu: 53.91 | total time: 62.74m | eta: 48.1m step 03157/05568 (56.70%) | loss: 2.513582 | lrm: 0.68 | dt: 1186.97ms | tok/sec: 883,409 | bf16_mfu: 54.27 | total time: 62.76m | eta: 48.1m step 03158/05568 (56.72%) | loss: 2.508943 | lrm: 0.68 | dt: 1192.01ms | tok/sec: 879,672 | bf16_mfu: 54.04 | total time: 62.78m | eta: 48.0m step 03159/05568 (56.73%) | loss: 2.503002 | lrm: 0.68 | dt: 1190.14ms | tok/sec: 881,049 | bf16_mfu: 54.12 | total time: 62.80m | eta: 48.0m step 03160/05568 (56.75%) | loss: 2.494038 | lrm: 0.68 | dt: 1192.27ms | tok/sec: 879,481 | bf16_mfu: 54.02 | total time: 62.82m | eta: 48.0m step 03161/05568 (56.77%) | loss: 2.502739 | lrm: 0.68 | dt: 1199.57ms | tok/sec: 874,125 | bf16_mfu: 53.70 | total time: 62.84m | eta: 48.0m step 03162/05568 (56.79%) | loss: 2.505067 | lrm: 0.68 | dt: 1187.43ms | tok/sec: 883,060 | bf16_mfu: 54.24 | total time: 62.86m | eta: 48.0m step 03163/05568 (56.81%) | loss: 2.503669 | lrm: 0.68 | dt: 1198.93ms | tok/sec: 874,593 | bf16_mfu: 53.72 | total time: 62.88m | eta: 47.9m step 03164/05568 (56.82%) | loss: 2.503063 | lrm: 0.68 | dt: 1193.67ms | tok/sec: 878,445 | bf16_mfu: 53.96 | total time: 62.90m | eta: 47.9m step 03165/05568 (56.84%) | loss: 2.501611 | lrm: 0.68 | dt: 1194.69ms | tok/sec: 877,696 | bf16_mfu: 53.91 | total time: 62.92m | eta: 47.9m step 03166/05568 (56.86%) | loss: 2.509489 | lrm: 0.68 | dt: 1192.25ms | tok/sec: 879,493 | bf16_mfu: 54.03 | total time: 62.94m | eta: 47.9m step 03167/05568 (56.88%) | loss: 2.519954 | lrm: 0.68 | dt: 1192.15ms | tok/sec: 879,568 | bf16_mfu: 54.03 | total time: 62.96m | eta: 47.9m step 03168/05568 (56.90%) | loss: 2.528632 | lrm: 0.68 | dt: 1185.18ms | tok/sec: 884,743 | bf16_mfu: 54.35 | total time: 62.98m | eta: 47.8m step 03169/05568 (56.91%) | loss: 2.530878 | lrm: 0.68 | dt: 1194.63ms | tok/sec: 877,741 | bf16_mfu: 53.92 | total time: 63.00m | eta: 47.8m step 03170/05568 (56.93%) | loss: 2.527550 | lrm: 0.68 | dt: 1191.23ms | tok/sec: 880,244 | bf16_mfu: 54.07 | total time: 63.02m | eta: 47.8m step 03171/05568 (56.95%) | loss: 2.526347 | lrm: 0.68 | dt: 1190.89ms | tok/sec: 880,495 | bf16_mfu: 54.09 | total time: 63.04m | eta: 47.8m step 03172/05568 (56.97%) | loss: 2.530774 | lrm: 0.68 | dt: 1194.11ms | tok/sec: 878,122 | bf16_mfu: 53.94 | total time: 63.06m | eta: 47.8m step 03173/05568 (56.99%) | loss: 2.529041 | lrm: 0.68 | dt: 1193.68ms | tok/sec: 878,436 | bf16_mfu: 53.96 | total time: 63.08m | eta: 47.7m step 03174/05568 (57.00%) | loss: 2.524431 | lrm: 0.68 | dt: 1191.04ms | tok/sec: 880,388 | bf16_mfu: 54.08 | total time: 63.10m | eta: 47.7m step 03175/05568 (57.02%) | loss: 2.516590 | lrm: 0.68 | dt: 1196.28ms | tok/sec: 876,529 | bf16_mfu: 53.84 | total time: 63.12m | eta: 47.7m step 03176/05568 (57.04%) | loss: 2.518492 | lrm: 0.68 | dt: 1193.11ms | tok/sec: 878,858 | bf16_mfu: 53.99 | total time: 63.14m | eta: 47.7m step 03177/05568 (57.06%) | loss: 2.516670 | lrm: 0.68 | dt: 1193.61ms | tok/sec: 878,488 | bf16_mfu: 53.96 | total time: 63.16m | eta: 47.7m step 03178/05568 (57.08%) | loss: 2.517519 | lrm: 0.68 | dt: 1193.59ms | tok/sec: 878,509 | bf16_mfu: 53.96 | total time: 63.18m | eta: 47.6m step 03179/05568 (57.09%) | loss: 2.511517 | lrm: 0.68 | dt: 1191.44ms | tok/sec: 880,088 | bf16_mfu: 54.06 | total time: 63.20m | eta: 47.6m step 03180/05568 (57.11%) | loss: 2.513790 | lrm: 0.68 | dt: 1190.10ms | tok/sec: 881,081 | bf16_mfu: 54.12 | total time: 63.22m | eta: 47.6m step 03181/05568 (57.13%) | loss: 2.518360 | lrm: 0.68 | dt: 1194.78ms | tok/sec: 877,628 | bf16_mfu: 53.91 | total time: 63.24m | eta: 47.6m step 03182/05568 (57.15%) | loss: 2.525054 | lrm: 0.68 | dt: 1191.13ms | tok/sec: 880,319 | bf16_mfu: 54.08 | total time: 63.26m | eta: 47.6m step 03183/05568 (57.17%) | loss: 2.521854 | lrm: 0.68 | dt: 1192.02ms | tok/sec: 879,662 | bf16_mfu: 54.04 | total time: 63.28m | eta: 47.5m step 03184/05568 (57.18%) | loss: 2.530109 | lrm: 0.68 | dt: 1193.98ms | tok/sec: 878,216 | bf16_mfu: 53.95 | total time: 63.30m | eta: 47.5m step 03185/05568 (57.20%) | loss: 2.538809 | lrm: 0.68 | dt: 1191.84ms | tok/sec: 879,796 | bf16_mfu: 54.04 | total time: 63.32m | eta: 47.5m step 03186/05568 (57.22%) | loss: 2.530162 | lrm: 0.68 | dt: 1194.55ms | tok/sec: 877,797 | bf16_mfu: 53.92 | total time: 63.34m | eta: 47.5m step 03187/05568 (57.24%) | loss: 2.525652 | lrm: 0.68 | dt: 1190.39ms | tok/sec: 880,868 | bf16_mfu: 54.11 | total time: 63.36m | eta: 47.5m step 03188/05568 (57.26%) | loss: 2.527692 | lrm: 0.67 | dt: 1199.10ms | tok/sec: 874,472 | bf16_mfu: 53.72 | total time: 63.38m | eta: 47.4m step 03189/05568 (57.27%) | loss: 2.521674 | lrm: 0.67 | dt: 1193.97ms | tok/sec: 878,229 | bf16_mfu: 53.95 | total time: 63.40m | eta: 47.4m step 03190/05568 (57.29%) | loss: 2.512725 | lrm: 0.67 | dt: 1191.45ms | tok/sec: 880,086 | bf16_mfu: 54.06 | total time: 63.42m | eta: 47.4m step 03191/05568 (57.31%) | loss: 2.510534 | lrm: 0.67 | dt: 1187.77ms | tok/sec: 882,808 | bf16_mfu: 54.23 | total time: 63.44m | eta: 47.4m step 03192/05568 (57.33%) | loss: 2.514731 | lrm: 0.67 | dt: 1192.74ms | tok/sec: 879,133 | bf16_mfu: 54.00 | total time: 63.46m | eta: 47.4m step 03193/05568 (57.35%) | loss: 2.513966 | lrm: 0.67 | dt: 1189.13ms | tok/sec: 881,803 | bf16_mfu: 54.17 | total time: 63.48m | eta: 47.3m step 03194/05568 (57.36%) | loss: 2.512587 | lrm: 0.67 | dt: 1195.09ms | tok/sec: 877,403 | bf16_mfu: 53.90 | total time: 63.50m | eta: 47.3m step 03195/05568 (57.38%) | loss: 2.497271 | lrm: 0.67 | dt: 1195.69ms | tok/sec: 876,963 | bf16_mfu: 53.87 | total time: 63.52m | eta: 47.3m step 03196/05568 (57.40%) | loss: 2.507331 | lrm: 0.67 | dt: 1188.61ms | tok/sec: 882,183 | bf16_mfu: 54.19 | total time: 63.54m | eta: 47.3m step 03197/05568 (57.42%) | loss: 2.519148 | lrm: 0.67 | dt: 1194.29ms | tok/sec: 877,990 | bf16_mfu: 53.93 | total time: 63.56m | eta: 47.3m step 03198/05568 (57.44%) | loss: 2.531156 | lrm: 0.67 | dt: 1195.06ms | tok/sec: 877,425 | bf16_mfu: 53.90 | total time: 63.58m | eta: 47.2m step 03199/05568 (57.45%) | loss: 2.524210 | lrm: 0.67 | dt: 1190.50ms | tok/sec: 880,784 | bf16_mfu: 54.10 | total time: 63.60m | eta: 47.2m step 03200/05568 (57.47%) | loss: 2.520674 | lrm: 0.67 | dt: 1189.22ms | tok/sec: 881,735 | bf16_mfu: 54.16 | total time: 63.62m | eta: 47.2m step 03201/05568 (57.49%) | loss: 2.526530 | lrm: 0.67 | dt: 1188.98ms | tok/sec: 881,912 | bf16_mfu: 54.17 | total time: 63.64m | eta: 47.2m step 03202/05568 (57.51%) | loss: 2.534179 | lrm: 0.67 | dt: 1193.90ms | tok/sec: 878,278 | bf16_mfu: 53.95 | total time: 63.66m | eta: 47.2m step 03203/05568 (57.53%) | loss: 2.527120 | lrm: 0.67 | dt: 1191.81ms | tok/sec: 879,814 | bf16_mfu: 54.05 | total time: 63.68m | eta: 47.1m step 03204/05568 (57.54%) | loss: 2.526294 | lrm: 0.67 | dt: 1194.32ms | tok/sec: 877,972 | bf16_mfu: 53.93 | total time: 63.70m | eta: 47.1m step 03205/05568 (57.56%) | loss: 2.532134 | lrm: 0.67 | dt: 1195.35ms | tok/sec: 877,211 | bf16_mfu: 53.89 | total time: 63.72m | eta: 47.1m step 03206/05568 (57.58%) | loss: 2.529912 | lrm: 0.67 | dt: 1188.31ms | tok/sec: 882,408 | bf16_mfu: 54.20 | total time: 63.74m | eta: 47.1m step 03207/05568 (57.60%) | loss: 2.534565 | lrm: 0.67 | dt: 1191.27ms | tok/sec: 880,214 | bf16_mfu: 54.07 | total time: 63.76m | eta: 47.1m step 03208/05568 (57.61%) | loss: 2.531451 | lrm: 0.67 | dt: 1191.38ms | tok/sec: 880,137 | bf16_mfu: 54.06 | total time: 63.78m | eta: 47.0m step 03209/05568 (57.63%) | loss: 2.526136 | lrm: 0.67 | dt: 1189.29ms | tok/sec: 881,681 | bf16_mfu: 54.16 | total time: 63.80m | eta: 47.0m step 03210/05568 (57.65%) | loss: 2.519535 | lrm: 0.67 | dt: 1195.21ms | tok/sec: 877,317 | bf16_mfu: 53.89 | total time: 63.82m | eta: 47.0m step 03211/05568 (57.67%) | loss: 2.518484 | lrm: 0.67 | dt: 1193.97ms | tok/sec: 878,222 | bf16_mfu: 53.95 | total time: 63.84m | eta: 47.0m step 03212/05568 (57.69%) | loss: 2.525257 | lrm: 0.67 | dt: 1194.69ms | tok/sec: 877,698 | bf16_mfu: 53.92 | total time: 63.86m | eta: 47.0m step 03213/05568 (57.70%) | loss: 2.515147 | lrm: 0.67 | dt: 1190.56ms | tok/sec: 880,745 | bf16_mfu: 54.10 | total time: 63.88m | eta: 46.9m step 03214/05568 (57.72%) | loss: 2.514953 | lrm: 0.67 | dt: 1194.02ms | tok/sec: 878,190 | bf16_mfu: 53.95 | total time: 63.90m | eta: 46.9m step 03215/05568 (57.74%) | loss: 2.511377 | lrm: 0.67 | dt: 1191.67ms | tok/sec: 879,920 | bf16_mfu: 54.05 | total time: 63.92m | eta: 46.9m step 03216/05568 (57.76%) | loss: 2.516344 | lrm: 0.67 | dt: 1197.61ms | tok/sec: 875,559 | bf16_mfu: 53.78 | total time: 63.94m | eta: 46.9m step 03217/05568 (57.78%) | loss: 2.509419 | lrm: 0.67 | dt: 1196.05ms | tok/sec: 876,699 | bf16_mfu: 53.85 | total time: 63.96m | eta: 46.9m step 03218/05568 (57.79%) | loss: 2.509015 | lrm: 0.67 | dt: 1191.33ms | tok/sec: 880,174 | bf16_mfu: 54.07 | total time: 63.98m | eta: 46.8m step 03219/05568 (57.81%) | loss: 2.506948 | lrm: 0.67 | dt: 1194.64ms | tok/sec: 877,732 | bf16_mfu: 53.92 | total time: 64.00m | eta: 46.8m step 03220/05568 (57.83%) | loss: 2.508675 | lrm: 0.67 | dt: 1190.28ms | tok/sec: 880,948 | bf16_mfu: 54.11 | total time: 64.02m | eta: 46.8m step 03221/05568 (57.85%) | loss: 2.498242 | lrm: 0.67 | dt: 1189.45ms | tok/sec: 881,560 | bf16_mfu: 54.15 | total time: 64.04m | eta: 46.8m step 03222/05568 (57.87%) | loss: 2.498049 | lrm: 0.67 | dt: 1189.81ms | tok/sec: 881,297 | bf16_mfu: 54.14 | total time: 64.06m | eta: 46.8m step 03223/05568 (57.88%) | loss: 2.496717 | lrm: 0.67 | dt: 1190.70ms | tok/sec: 880,641 | bf16_mfu: 54.10 | total time: 64.08m | eta: 46.7m step 03224/05568 (57.90%) | loss: 2.503264 | lrm: 0.67 | dt: 1191.33ms | tok/sec: 880,174 | bf16_mfu: 54.07 | total time: 64.10m | eta: 46.7m step 03225/05568 (57.92%) | loss: 2.511777 | lrm: 0.67 | dt: 1190.71ms | tok/sec: 880,629 | bf16_mfu: 54.10 | total time: 64.12m | eta: 46.7m step 03226/05568 (57.94%) | loss: 2.521450 | lrm: 0.66 | dt: 1196.02ms | tok/sec: 876,718 | bf16_mfu: 53.85 | total time: 64.14m | eta: 46.7m step 03227/05568 (57.96%) | loss: 2.523260 | lrm: 0.66 | dt: 1194.04ms | tok/sec: 878,171 | bf16_mfu: 53.94 | total time: 64.16m | eta: 46.7m step 03228/05568 (57.97%) | loss: 2.530782 | lrm: 0.66 | dt: 1189.46ms | tok/sec: 881,554 | bf16_mfu: 54.15 | total time: 64.18m | eta: 46.6m step 03229/05568 (57.99%) | loss: 2.524919 | lrm: 0.66 | dt: 1191.80ms | tok/sec: 879,826 | bf16_mfu: 54.05 | total time: 64.20m | eta: 46.6m step 03230/05568 (58.01%) | loss: 2.519713 | lrm: 0.66 | dt: 1194.93ms | tok/sec: 877,522 | bf16_mfu: 53.90 | total time: 64.22m | eta: 46.6m step 03231/05568 (58.03%) | loss: 2.524538 | lrm: 0.66 | dt: 1188.48ms | tok/sec: 882,282 | bf16_mfu: 54.20 | total time: 64.24m | eta: 46.6m step 03232/05568 (58.05%) | loss: 2.527892 | lrm: 0.66 | dt: 1195.31ms | tok/sec: 877,244 | bf16_mfu: 53.89 | total time: 64.26m | eta: 46.6m step 03233/05568 (58.06%) | loss: 2.526505 | lrm: 0.66 | dt: 1188.53ms | tok/sec: 882,246 | bf16_mfu: 54.19 | total time: 64.27m | eta: 46.5m step 03234/05568 (58.08%) | loss: 2.528003 | lrm: 0.66 | dt: 1188.75ms | tok/sec: 882,083 | bf16_mfu: 54.18 | total time: 64.29m | eta: 46.5m step 03235/05568 (58.10%) | loss: 2.522890 | lrm: 0.66 | dt: 1190.51ms | tok/sec: 880,782 | bf16_mfu: 54.10 | total time: 64.31m | eta: 46.5m step 03236/05568 (58.12%) | loss: 2.534566 | lrm: 0.66 | dt: 1192.19ms | tok/sec: 879,538 | bf16_mfu: 54.03 | total time: 64.33m | eta: 46.5m step 03237/05568 (58.14%) | loss: 2.540856 | lrm: 0.66 | dt: 1194.98ms | tok/sec: 877,482 | bf16_mfu: 53.90 | total time: 64.35m | eta: 46.5m step 03238/05568 (58.15%) | loss: 2.540109 | lrm: 0.66 | dt: 1192.35ms | tok/sec: 879,419 | bf16_mfu: 54.02 | total time: 64.37m | eta: 46.4m step 03239/05568 (58.17%) | loss: 2.532933 | lrm: 0.66 | dt: 1192.49ms | tok/sec: 879,313 | bf16_mfu: 54.01 | total time: 64.39m | eta: 46.4m step 03240/05568 (58.19%) | loss: 2.530021 | lrm: 0.66 | dt: 1193.65ms | tok/sec: 878,460 | bf16_mfu: 53.96 | total time: 64.41m | eta: 46.4m step 03241/05568 (58.21%) | loss: 2.524720 | lrm: 0.66 | dt: 1192.50ms | tok/sec: 879,309 | bf16_mfu: 54.01 | total time: 64.43m | eta: 46.4m step 03242/05568 (58.23%) | loss: 2.523743 | lrm: 0.66 | dt: 1191.68ms | tok/sec: 879,912 | bf16_mfu: 54.05 | total time: 64.45m | eta: 46.4m step 03243/05568 (58.24%) | loss: 2.525196 | lrm: 0.66 | dt: 1192.41ms | tok/sec: 879,372 | bf16_mfu: 54.02 | total time: 64.47m | eta: 46.3m step 03244/05568 (58.26%) | loss: 2.534641 | lrm: 0.66 | dt: 1194.21ms | tok/sec: 878,053 | bf16_mfu: 53.94 | total time: 64.49m | eta: 46.3m step 03245/05568 (58.28%) | loss: 2.531233 | lrm: 0.66 | dt: 1194.40ms | tok/sec: 877,912 | bf16_mfu: 53.93 | total time: 64.51m | eta: 46.3m step 03246/05568 (58.30%) | loss: 2.537660 | lrm: 0.66 | dt: 1191.77ms | tok/sec: 879,848 | bf16_mfu: 54.05 | total time: 64.53m | eta: 46.3m step 03247/05568 (58.32%) | loss: 2.541974 | lrm: 0.66 | dt: 1190.47ms | tok/sec: 880,805 | bf16_mfu: 54.11 | total time: 64.55m | eta: 46.3m step 03248/05568 (58.33%) | loss: 2.531445 | lrm: 0.66 | dt: 1192.88ms | tok/sec: 879,032 | bf16_mfu: 54.00 | total time: 64.57m | eta: 46.2m step 03249/05568 (58.35%) | loss: 2.530557 | lrm: 0.66 | dt: 1190.37ms | tok/sec: 880,882 | bf16_mfu: 54.11 | total time: 64.59m | eta: 46.2m step:3250/5568 val_bpb:0.766625 val_time:3.90s step 03250/05568 (58.37%) | loss: 2.532988 | lrm: 0.66 | dt: 1196.36ms | tok/sec: 876,470 | bf16_mfu: 53.84 | total time: 64.61m | eta: 46.2m step 03251/05568 (58.39%) | loss: 2.532865 | lrm: 0.66 | dt: 1194.54ms | tok/sec: 877,804 | bf16_mfu: 53.92 | total time: 64.63m | eta: 46.2m step 03252/05568 (58.41%) | loss: 2.529914 | lrm: 0.66 | dt: 1192.15ms | tok/sec: 879,564 | bf16_mfu: 54.03 | total time: 64.65m | eta: 46.2m step 03253/05568 (58.42%) | loss: 2.521998 | lrm: 0.66 | dt: 1191.34ms | tok/sec: 880,167 | bf16_mfu: 54.07 | total time: 64.67m | eta: 46.1m step 03254/05568 (58.44%) | loss: 2.510169 | lrm: 0.66 | dt: 1196.07ms | tok/sec: 876,687 | bf16_mfu: 53.85 | total time: 64.69m | eta: 46.1m step 03255/05568 (58.46%) | loss: 2.506536 | lrm: 0.66 | dt: 1190.87ms | tok/sec: 880,511 | bf16_mfu: 54.09 | total time: 64.71m | eta: 46.1m step 03256/05568 (58.48%) | loss: 2.504505 | lrm: 0.66 | dt: 1194.09ms | tok/sec: 878,137 | bf16_mfu: 53.94 | total time: 64.73m | eta: 46.1m step 03257/05568 (58.49%) | loss: 2.511578 | lrm: 0.66 | dt: 1193.35ms | tok/sec: 878,683 | bf16_mfu: 53.98 | total time: 64.75m | eta: 46.1m step 03258/05568 (58.51%) | loss: 2.512048 | lrm: 0.66 | dt: 1191.40ms | tok/sec: 880,119 | bf16_mfu: 54.06 | total time: 64.77m | eta: 46.0m step 03259/05568 (58.53%) | loss: 2.513959 | lrm: 0.66 | dt: 1194.84ms | tok/sec: 877,586 | bf16_mfu: 53.91 | total time: 64.79m | eta: 46.0m step 03260/05568 (58.55%) | loss: 2.519298 | lrm: 0.66 | dt: 1191.36ms | tok/sec: 880,154 | bf16_mfu: 54.07 | total time: 64.81m | eta: 46.0m step 03261/05568 (58.57%) | loss: 2.512423 | lrm: 0.66 | dt: 1189.38ms | tok/sec: 881,618 | bf16_mfu: 54.16 | total time: 64.83m | eta: 46.0m step 03262/05568 (58.58%) | loss: 2.526278 | lrm: 0.66 | dt: 1187.81ms | tok/sec: 882,779 | bf16_mfu: 54.23 | total time: 64.85m | eta: 46.0m step 03263/05568 (58.60%) | loss: 2.528145 | lrm: 0.66 | dt: 1192.08ms | tok/sec: 879,620 | bf16_mfu: 54.03 | total time: 64.87m | eta: 45.9m step 03264/05568 (58.62%) | loss: 2.529731 | lrm: 0.65 | dt: 1191.22ms | tok/sec: 880,251 | bf16_mfu: 54.07 | total time: 64.89m | eta: 45.9m step 03265/05568 (58.64%) | loss: 2.528847 | lrm: 0.65 | dt: 1191.49ms | tok/sec: 880,056 | bf16_mfu: 54.06 | total time: 64.91m | eta: 45.9m step 03266/05568 (58.66%) | loss: 2.538851 | lrm: 0.65 | dt: 1197.10ms | tok/sec: 875,931 | bf16_mfu: 53.81 | total time: 64.93m | eta: 45.9m step 03267/05568 (58.67%) | loss: 2.544557 | lrm: 0.65 | dt: 1192.99ms | tok/sec: 878,945 | bf16_mfu: 53.99 | total time: 64.95m | eta: 45.9m step 03268/05568 (58.69%) | loss: 2.530893 | lrm: 0.65 | dt: 1189.37ms | tok/sec: 881,622 | bf16_mfu: 54.16 | total time: 64.97m | eta: 45.8m step 03269/05568 (58.71%) | loss: 2.535308 | lrm: 0.65 | dt: 1191.05ms | tok/sec: 880,382 | bf16_mfu: 54.08 | total time: 64.99m | eta: 45.8m step 03270/05568 (58.73%) | loss: 2.525847 | lrm: 0.65 | dt: 1190.22ms | tok/sec: 880,994 | bf16_mfu: 54.12 | total time: 65.01m | eta: 45.8m step 03271/05568 (58.75%) | loss: 2.538089 | lrm: 0.65 | dt: 1195.12ms | tok/sec: 877,381 | bf16_mfu: 53.90 | total time: 65.03m | eta: 45.8m step 03272/05568 (58.76%) | loss: 2.538645 | lrm: 0.65 | dt: 1187.79ms | tok/sec: 882,795 | bf16_mfu: 54.23 | total time: 65.05m | eta: 45.8m step 03273/05568 (58.78%) | loss: 2.526654 | lrm: 0.65 | dt: 1190.67ms | tok/sec: 880,661 | bf16_mfu: 54.10 | total time: 65.07m | eta: 45.7m step 03274/05568 (58.80%) | loss: 2.526006 | lrm: 0.65 | dt: 1195.21ms | tok/sec: 877,316 | bf16_mfu: 53.89 | total time: 65.09m | eta: 45.7m step 03275/05568 (58.82%) | loss: 2.523468 | lrm: 0.65 | dt: 1192.15ms | tok/sec: 879,567 | bf16_mfu: 54.03 | total time: 65.11m | eta: 45.7m step 03276/05568 (58.84%) | loss: 2.523422 | lrm: 0.65 | dt: 1194.38ms | tok/sec: 877,928 | bf16_mfu: 53.93 | total time: 65.13m | eta: 45.7m step 03277/05568 (58.85%) | loss: 2.535493 | lrm: 0.65 | dt: 1189.37ms | tok/sec: 881,622 | bf16_mfu: 54.16 | total time: 65.15m | eta: 45.7m step 03278/05568 (58.87%) | loss: 2.543532 | lrm: 0.65 | dt: 1192.78ms | tok/sec: 879,104 | bf16_mfu: 54.00 | total time: 65.17m | eta: 45.6m step 03279/05568 (58.89%) | loss: 2.537126 | lrm: 0.65 | dt: 1196.36ms | tok/sec: 876,471 | bf16_mfu: 53.84 | total time: 65.19m | eta: 45.6m step 03280/05568 (58.91%) | loss: 2.529193 | lrm: 0.65 | dt: 1195.67ms | tok/sec: 876,979 | bf16_mfu: 53.87 | total time: 65.21m | eta: 45.6m step 03281/05568 (58.93%) | loss: 2.545321 | lrm: 0.65 | dt: 1191.92ms | tok/sec: 879,734 | bf16_mfu: 54.04 | total time: 65.23m | eta: 45.6m step 03282/05568 (58.94%) | loss: 2.539919 | lrm: 0.65 | dt: 1189.96ms | tok/sec: 881,188 | bf16_mfu: 54.13 | total time: 65.25m | eta: 45.6m step 03283/05568 (58.96%) | loss: 2.533284 | lrm: 0.65 | dt: 1195.62ms | tok/sec: 877,014 | bf16_mfu: 53.87 | total time: 65.27m | eta: 45.5m step 03284/05568 (58.98%) | loss: 2.529966 | lrm: 0.65 | dt: 1190.88ms | tok/sec: 880,503 | bf16_mfu: 54.09 | total time: 65.29m | eta: 45.5m step 03285/05568 (59.00%) | loss: 2.533197 | lrm: 0.65 | dt: 1197.35ms | tok/sec: 875,744 | bf16_mfu: 53.80 | total time: 65.31m | eta: 45.5m step 03286/05568 (59.02%) | loss: 2.531760 | lrm: 0.65 | dt: 1193.47ms | tok/sec: 878,593 | bf16_mfu: 53.97 | total time: 65.33m | eta: 45.5m step 03287/05568 (59.03%) | loss: 2.541149 | lrm: 0.65 | dt: 1188.28ms | tok/sec: 882,433 | bf16_mfu: 54.21 | total time: 65.35m | eta: 45.5m step 03288/05568 (59.05%) | loss: 2.539532 | lrm: 0.65 | dt: 1193.05ms | tok/sec: 878,905 | bf16_mfu: 53.99 | total time: 65.37m | eta: 45.4m step 03289/05568 (59.07%) | loss: 2.533558 | lrm: 0.65 | dt: 1189.73ms | tok/sec: 881,356 | bf16_mfu: 54.14 | total time: 65.39m | eta: 45.4m step 03290/05568 (59.09%) | loss: 2.530940 | lrm: 0.65 | dt: 1191.94ms | tok/sec: 879,725 | bf16_mfu: 54.04 | total time: 65.41m | eta: 45.4m step 03291/05568 (59.11%) | loss: 2.535983 | lrm: 0.65 | dt: 1192.98ms | tok/sec: 878,951 | bf16_mfu: 53.99 | total time: 65.43m | eta: 45.4m step 03292/05568 (59.12%) | loss: 2.529232 | lrm: 0.65 | dt: 1192.97ms | tok/sec: 878,965 | bf16_mfu: 53.99 | total time: 65.45m | eta: 45.4m step 03293/05568 (59.14%) | loss: 2.529334 | lrm: 0.65 | dt: 1194.72ms | tok/sec: 877,672 | bf16_mfu: 53.91 | total time: 65.47m | eta: 45.3m step 03294/05568 (59.16%) | loss: 2.530290 | lrm: 0.65 | dt: 1191.01ms | tok/sec: 880,407 | bf16_mfu: 54.08 | total time: 65.49m | eta: 45.3m step 03295/05568 (59.18%) | loss: 2.526991 | lrm: 0.65 | dt: 1192.10ms | tok/sec: 879,604 | bf16_mfu: 54.03 | total time: 65.51m | eta: 45.3m step 03296/05568 (59.20%) | loss: 2.520762 | lrm: 0.65 | dt: 1189.38ms | tok/sec: 881,619 | bf16_mfu: 54.16 | total time: 65.53m | eta: 45.3m step 03297/05568 (59.21%) | loss: 2.534814 | lrm: 0.65 | dt: 1190.49ms | tok/sec: 880,797 | bf16_mfu: 54.11 | total time: 65.55m | eta: 45.3m step 03298/05568 (59.23%) | loss: 2.526601 | lrm: 0.65 | dt: 1193.31ms | tok/sec: 878,708 | bf16_mfu: 53.98 | total time: 65.57m | eta: 45.2m step 03299/05568 (59.25%) | loss: 2.530671 | lrm: 0.65 | dt: 1189.91ms | tok/sec: 881,222 | bf16_mfu: 54.13 | total time: 65.59m | eta: 45.2m step 03300/05568 (59.27%) | loss: 2.528266 | lrm: 0.65 | dt: 1196.78ms | tok/sec: 876,166 | bf16_mfu: 53.82 | total time: 65.61m | eta: 45.2m step 03301/05568 (59.29%) | loss: 2.521536 | lrm: 0.65 | dt: 1189.57ms | tok/sec: 881,473 | bf16_mfu: 54.15 | total time: 65.63m | eta: 45.2m step 03302/05568 (59.30%) | loss: 2.523762 | lrm: 0.64 | dt: 1196.85ms | tok/sec: 876,109 | bf16_mfu: 53.82 | total time: 65.65m | eta: 45.2m step 03303/05568 (59.32%) | loss: 2.535708 | lrm: 0.64 | dt: 1190.92ms | tok/sec: 880,477 | bf16_mfu: 54.09 | total time: 65.67m | eta: 45.1m step 03304/05568 (59.34%) | loss: 2.535872 | lrm: 0.64 | dt: 1194.99ms | tok/sec: 877,478 | bf16_mfu: 53.90 | total time: 65.69m | eta: 45.1m step 03305/05568 (59.36%) | loss: 2.533665 | lrm: 0.64 | dt: 1190.94ms | tok/sec: 880,463 | bf16_mfu: 54.08 | total time: 65.71m | eta: 45.1m step 03306/05568 (59.38%) | loss: 2.512580 | lrm: 0.64 | dt: 1191.02ms | tok/sec: 880,400 | bf16_mfu: 54.08 | total time: 65.73m | eta: 45.1m step 03307/05568 (59.39%) | loss: 2.514505 | lrm: 0.64 | dt: 1196.69ms | tok/sec: 876,227 | bf16_mfu: 53.82 | total time: 65.75m | eta: 45.1m step 03308/05568 (59.41%) | loss: 2.504328 | lrm: 0.64 | dt: 1197.90ms | tok/sec: 875,346 | bf16_mfu: 53.77 | total time: 65.77m | eta: 45.0m step 03309/05568 (59.43%) | loss: 2.511455 | lrm: 0.64 | dt: 1191.94ms | tok/sec: 879,724 | bf16_mfu: 54.04 | total time: 65.79m | eta: 45.0m step 03310/05568 (59.45%) | loss: 2.509268 | lrm: 0.64 | dt: 1192.50ms | tok/sec: 879,310 | bf16_mfu: 54.01 | total time: 65.81m | eta: 45.0m step 03311/05568 (59.46%) | loss: 2.516315 | lrm: 0.64 | dt: 1198.85ms | tok/sec: 874,648 | bf16_mfu: 53.73 | total time: 65.83m | eta: 45.0m step 03312/05568 (59.48%) | loss: 2.515206 | lrm: 0.64 | dt: 1190.97ms | tok/sec: 880,439 | bf16_mfu: 54.08 | total time: 65.85m | eta: 45.0m step 03313/05568 (59.50%) | loss: 2.524252 | lrm: 0.64 | dt: 1189.60ms | tok/sec: 881,454 | bf16_mfu: 54.15 | total time: 65.86m | eta: 44.9m step 03314/05568 (59.52%) | loss: 2.532181 | lrm: 0.64 | dt: 1192.36ms | tok/sec: 879,411 | bf16_mfu: 54.02 | total time: 65.88m | eta: 44.9m step 03315/05568 (59.54%) | loss: 2.530153 | lrm: 0.64 | dt: 1191.42ms | tok/sec: 880,109 | bf16_mfu: 54.06 | total time: 65.90m | eta: 44.9m step 03316/05568 (59.55%) | loss: 2.529669 | lrm: 0.64 | dt: 1192.17ms | tok/sec: 879,553 | bf16_mfu: 54.03 | total time: 65.92m | eta: 44.9m step 03317/05568 (59.57%) | loss: 2.532458 | lrm: 0.64 | dt: 1195.34ms | tok/sec: 877,216 | bf16_mfu: 53.89 | total time: 65.94m | eta: 44.9m step 03318/05568 (59.59%) | loss: 2.528857 | lrm: 0.64 | dt: 1190.79ms | tok/sec: 880,575 | bf16_mfu: 54.09 | total time: 65.96m | eta: 44.8m step 03319/05568 (59.61%) | loss: 2.535683 | lrm: 0.64 | dt: 1192.37ms | tok/sec: 879,406 | bf16_mfu: 54.02 | total time: 65.98m | eta: 44.8m step 03320/05568 (59.63%) | loss: 2.535689 | lrm: 0.64 | dt: 1191.27ms | tok/sec: 880,219 | bf16_mfu: 54.07 | total time: 66.00m | eta: 44.8m step 03321/05568 (59.64%) | loss: 2.526026 | lrm: 0.64 | dt: 1191.64ms | tok/sec: 879,944 | bf16_mfu: 54.05 | total time: 66.02m | eta: 44.8m step 03322/05568 (59.66%) | loss: 2.513391 | lrm: 0.64 | dt: 1187.81ms | tok/sec: 882,782 | bf16_mfu: 54.23 | total time: 66.04m | eta: 44.8m step 03323/05568 (59.68%) | loss: 2.511903 | lrm: 0.64 | dt: 1192.71ms | tok/sec: 879,153 | bf16_mfu: 54.00 | total time: 66.06m | eta: 44.7m step 03324/05568 (59.70%) | loss: 2.509816 | lrm: 0.64 | dt: 1188.47ms | tok/sec: 882,292 | bf16_mfu: 54.20 | total time: 66.08m | eta: 44.7m step 03325/05568 (59.72%) | loss: 2.510251 | lrm: 0.64 | dt: 1191.17ms | tok/sec: 880,289 | bf16_mfu: 54.07 | total time: 66.10m | eta: 44.7m step 03326/05568 (59.73%) | loss: 2.506742 | lrm: 0.64 | dt: 1195.93ms | tok/sec: 876,787 | bf16_mfu: 53.86 | total time: 66.12m | eta: 44.7m step 03327/05568 (59.75%) | loss: 2.512064 | lrm: 0.64 | dt: 1187.51ms | tok/sec: 883,005 | bf16_mfu: 54.24 | total time: 66.14m | eta: 44.7m step 03328/05568 (59.77%) | loss: 2.491218 | lrm: 0.64 | dt: 1195.53ms | tok/sec: 877,077 | bf16_mfu: 53.88 | total time: 66.16m | eta: 44.6m step 03329/05568 (59.79%) | loss: 2.496361 | lrm: 0.64 | dt: 1189.46ms | tok/sec: 881,555 | bf16_mfu: 54.15 | total time: 66.18m | eta: 44.6m step 03330/05568 (59.81%) | loss: 2.496605 | lrm: 0.64 | dt: 1194.79ms | tok/sec: 877,627 | bf16_mfu: 53.91 | total time: 66.20m | eta: 44.6m step 03331/05568 (59.82%) | loss: 2.505283 | lrm: 0.64 | dt: 1194.03ms | tok/sec: 878,183 | bf16_mfu: 53.94 | total time: 66.22m | eta: 44.6m step 03332/05568 (59.84%) | loss: 2.505112 | lrm: 0.64 | dt: 1189.52ms | tok/sec: 881,511 | bf16_mfu: 54.15 | total time: 66.24m | eta: 44.6m step 03333/05568 (59.86%) | loss: 2.510884 | lrm: 0.64 | dt: 1192.56ms | tok/sec: 879,265 | bf16_mfu: 54.01 | total time: 66.26m | eta: 44.5m step 03334/05568 (59.88%) | loss: 2.512036 | lrm: 0.64 | dt: 1193.26ms | tok/sec: 878,752 | bf16_mfu: 53.98 | total time: 66.28m | eta: 44.5m step 03335/05568 (59.90%) | loss: 2.515189 | lrm: 0.64 | dt: 1195.33ms | tok/sec: 877,229 | bf16_mfu: 53.89 | total time: 66.30m | eta: 44.5m step 03336/05568 (59.91%) | loss: 2.512991 | lrm: 0.64 | dt: 1197.43ms | tok/sec: 875,691 | bf16_mfu: 53.79 | total time: 66.32m | eta: 44.5m step 03337/05568 (59.93%) | loss: 2.524424 | lrm: 0.64 | dt: 1199.19ms | tok/sec: 874,402 | bf16_mfu: 53.71 | total time: 66.34m | eta: 44.5m step 03338/05568 (59.95%) | loss: 2.514201 | lrm: 0.64 | dt: 1192.13ms | tok/sec: 879,579 | bf16_mfu: 54.03 | total time: 66.36m | eta: 44.4m step 03339/05568 (59.97%) | loss: 2.507591 | lrm: 0.64 | dt: 1189.21ms | tok/sec: 881,744 | bf16_mfu: 54.16 | total time: 66.38m | eta: 44.4m step 03340/05568 (59.99%) | loss: 2.509335 | lrm: 0.63 | dt: 1196.83ms | tok/sec: 876,130 | bf16_mfu: 53.82 | total time: 66.40m | eta: 44.4m step 03341/05568 (60.00%) | loss: 2.515248 | lrm: 0.63 | dt: 1194.33ms | tok/sec: 877,958 | bf16_mfu: 53.93 | total time: 66.42m | eta: 44.4m step 03342/05568 (60.02%) | loss: 2.514819 | lrm: 0.63 | dt: 1189.56ms | tok/sec: 881,483 | bf16_mfu: 54.15 | total time: 66.44m | eta: 44.4m step 03343/05568 (60.04%) | loss: 2.516393 | lrm: 0.63 | dt: 1194.59ms | tok/sec: 877,771 | bf16_mfu: 53.92 | total time: 66.46m | eta: 44.3m step 03344/05568 (60.06%) | loss: 2.520566 | lrm: 0.63 | dt: 1194.20ms | tok/sec: 878,056 | bf16_mfu: 53.94 | total time: 66.48m | eta: 44.3m step 03345/05568 (60.08%) | loss: 2.508707 | lrm: 0.63 | dt: 1188.05ms | tok/sec: 882,603 | bf16_mfu: 54.22 | total time: 66.50m | eta: 44.3m step 03346/05568 (60.09%) | loss: 2.504865 | lrm: 0.63 | dt: 1194.53ms | tok/sec: 877,814 | bf16_mfu: 53.92 | total time: 66.52m | eta: 44.3m step 03347/05568 (60.11%) | loss: 2.494822 | lrm: 0.63 | dt: 1191.58ms | tok/sec: 879,989 | bf16_mfu: 54.06 | total time: 66.54m | eta: 44.3m step 03348/05568 (60.13%) | loss: 2.498309 | lrm: 0.63 | dt: 1194.93ms | tok/sec: 877,524 | bf16_mfu: 53.90 | total time: 66.56m | eta: 44.2m step 03349/05568 (60.15%) | loss: 2.489606 | lrm: 0.63 | dt: 1194.84ms | tok/sec: 877,588 | bf16_mfu: 53.91 | total time: 66.58m | eta: 44.2m step 03350/05568 (60.17%) | loss: 2.487816 | lrm: 0.63 | dt: 1196.92ms | tok/sec: 876,060 | bf16_mfu: 53.81 | total time: 66.60m | eta: 44.2m step 03351/05568 (60.18%) | loss: 2.486929 | lrm: 0.63 | dt: 1190.79ms | tok/sec: 880,569 | bf16_mfu: 54.09 | total time: 66.62m | eta: 44.2m step 03352/05568 (60.20%) | loss: 2.489149 | lrm: 0.63 | dt: 1193.70ms | tok/sec: 878,428 | bf16_mfu: 53.96 | total time: 66.64m | eta: 44.2m step 03353/05568 (60.22%) | loss: 2.484249 | lrm: 0.63 | dt: 1194.15ms | tok/sec: 878,093 | bf16_mfu: 53.94 | total time: 66.66m | eta: 44.1m step 03354/05568 (60.24%) | loss: 2.487557 | lrm: 0.63 | dt: 1193.19ms | tok/sec: 878,798 | bf16_mfu: 53.98 | total time: 66.68m | eta: 44.1m step 03355/05568 (60.26%) | loss: 2.494825 | lrm: 0.63 | dt: 1189.01ms | tok/sec: 881,888 | bf16_mfu: 54.17 | total time: 66.70m | eta: 44.1m step 03356/05568 (60.27%) | loss: 2.498400 | lrm: 0.63 | dt: 1193.13ms | tok/sec: 878,847 | bf16_mfu: 53.99 | total time: 66.72m | eta: 44.1m step 03357/05568 (60.29%) | loss: 2.501747 | lrm: 0.63 | dt: 1191.42ms | tok/sec: 880,107 | bf16_mfu: 54.06 | total time: 66.74m | eta: 44.1m step 03358/05568 (60.31%) | loss: 2.493334 | lrm: 0.63 | dt: 1193.43ms | tok/sec: 878,624 | bf16_mfu: 53.97 | total time: 66.76m | eta: 44.0m step 03359/05568 (60.33%) | loss: 2.493146 | lrm: 0.63 | dt: 1193.50ms | tok/sec: 878,573 | bf16_mfu: 53.97 | total time: 66.78m | eta: 44.0m step 03360/05568 (60.34%) | loss: 2.498326 | lrm: 0.63 | dt: 1191.69ms | tok/sec: 879,905 | bf16_mfu: 54.05 | total time: 66.80m | eta: 44.0m step 03361/05568 (60.36%) | loss: 2.492306 | lrm: 0.63 | dt: 1193.23ms | tok/sec: 878,773 | bf16_mfu: 53.98 | total time: 66.82m | eta: 44.0m step 03362/05568 (60.38%) | loss: 2.483982 | lrm: 0.63 | dt: 1189.81ms | tok/sec: 881,294 | bf16_mfu: 54.14 | total time: 66.84m | eta: 44.0m step 03363/05568 (60.40%) | loss: 2.495925 | lrm: 0.63 | dt: 1194.51ms | tok/sec: 877,825 | bf16_mfu: 53.92 | total time: 66.86m | eta: 43.9m step 03364/05568 (60.42%) | loss: 2.498321 | lrm: 0.63 | dt: 1193.81ms | tok/sec: 878,346 | bf16_mfu: 53.95 | total time: 66.88m | eta: 43.9m step 03365/05568 (60.43%) | loss: 2.497854 | lrm: 0.63 | dt: 1196.49ms | tok/sec: 876,373 | bf16_mfu: 53.83 | total time: 66.90m | eta: 43.9m step 03366/05568 (60.45%) | loss: 2.499984 | lrm: 0.63 | dt: 1188.23ms | tok/sec: 882,469 | bf16_mfu: 54.21 | total time: 66.92m | eta: 43.9m step 03367/05568 (60.47%) | loss: 2.500905 | lrm: 0.63 | dt: 1192.13ms | tok/sec: 879,583 | bf16_mfu: 54.03 | total time: 66.94m | eta: 43.9m step 03368/05568 (60.49%) | loss: 2.495974 | lrm: 0.63 | dt: 1191.54ms | tok/sec: 880,018 | bf16_mfu: 54.06 | total time: 66.96m | eta: 43.8m step 03369/05568 (60.51%) | loss: 2.504303 | lrm: 0.63 | dt: 1193.77ms | tok/sec: 878,372 | bf16_mfu: 53.96 | total time: 66.98m | eta: 43.8m step 03370/05568 (60.52%) | loss: 2.509695 | lrm: 0.63 | dt: 1192.08ms | tok/sec: 879,615 | bf16_mfu: 54.03 | total time: 67.00m | eta: 43.8m step 03371/05568 (60.54%) | loss: 2.507574 | lrm: 0.63 | dt: 1193.91ms | tok/sec: 878,273 | bf16_mfu: 53.95 | total time: 67.02m | eta: 43.8m step 03372/05568 (60.56%) | loss: 2.512597 | lrm: 0.63 | dt: 1194.88ms | tok/sec: 877,557 | bf16_mfu: 53.91 | total time: 67.04m | eta: 43.8m step 03373/05568 (60.58%) | loss: 2.518674 | lrm: 0.63 | dt: 1190.04ms | tok/sec: 881,126 | bf16_mfu: 54.13 | total time: 67.06m | eta: 43.7m step 03374/05568 (60.60%) | loss: 2.517312 | lrm: 0.63 | dt: 1189.75ms | tok/sec: 881,341 | bf16_mfu: 54.14 | total time: 67.08m | eta: 43.7m step 03375/05568 (60.61%) | loss: 2.520049 | lrm: 0.63 | dt: 1193.56ms | tok/sec: 878,527 | bf16_mfu: 53.97 | total time: 67.10m | eta: 43.7m step 03376/05568 (60.63%) | loss: 2.524276 | lrm: 0.63 | dt: 1193.66ms | tok/sec: 878,455 | bf16_mfu: 53.96 | total time: 67.12m | eta: 43.7m step 03377/05568 (60.65%) | loss: 2.522111 | lrm: 0.63 | dt: 1191.10ms | tok/sec: 880,340 | bf16_mfu: 54.08 | total time: 67.14m | eta: 43.7m step 03378/05568 (60.67%) | loss: 2.518945 | lrm: 0.62 | dt: 1193.49ms | tok/sec: 878,579 | bf16_mfu: 53.97 | total time: 67.16m | eta: 43.6m step 03379/05568 (60.69%) | loss: 2.506986 | lrm: 0.62 | dt: 1195.56ms | tok/sec: 877,061 | bf16_mfu: 53.88 | total time: 67.18m | eta: 43.6m step 03380/05568 (60.70%) | loss: 2.520144 | lrm: 0.62 | dt: 1200.59ms | tok/sec: 873,383 | bf16_mfu: 53.65 | total time: 67.20m | eta: 43.6m step 03381/05568 (60.72%) | loss: 2.525287 | lrm: 0.62 | dt: 1188.72ms | tok/sec: 882,102 | bf16_mfu: 54.19 | total time: 67.22m | eta: 43.6m step 03382/05568 (60.74%) | loss: 2.520337 | lrm: 0.62 | dt: 1196.10ms | tok/sec: 876,659 | bf16_mfu: 53.85 | total time: 67.24m | eta: 43.6m step 03383/05568 (60.76%) | loss: 2.523789 | lrm: 0.62 | dt: 1192.67ms | tok/sec: 879,185 | bf16_mfu: 54.01 | total time: 67.26m | eta: 43.5m step 03384/05568 (60.78%) | loss: 2.511981 | lrm: 0.62 | dt: 1192.60ms | tok/sec: 879,235 | bf16_mfu: 54.01 | total time: 67.28m | eta: 43.5m step 03385/05568 (60.79%) | loss: 2.520122 | lrm: 0.62 | dt: 1196.33ms | tok/sec: 876,495 | bf16_mfu: 53.84 | total time: 67.30m | eta: 43.5m step 03386/05568 (60.81%) | loss: 2.514256 | lrm: 0.62 | dt: 1192.70ms | tok/sec: 879,159 | bf16_mfu: 54.00 | total time: 67.32m | eta: 43.5m step 03387/05568 (60.83%) | loss: 2.520255 | lrm: 0.62 | dt: 1192.51ms | tok/sec: 879,300 | bf16_mfu: 54.01 | total time: 67.34m | eta: 43.5m step 03388/05568 (60.85%) | loss: 2.525646 | lrm: 0.62 | dt: 1194.40ms | tok/sec: 877,909 | bf16_mfu: 53.93 | total time: 67.36m | eta: 43.4m step 03389/05568 (60.87%) | loss: 2.525041 | lrm: 0.62 | dt: 1187.73ms | tok/sec: 882,841 | bf16_mfu: 54.23 | total time: 67.38m | eta: 43.4m step 03390/05568 (60.88%) | loss: 2.515710 | lrm: 0.62 | dt: 1197.15ms | tok/sec: 875,890 | bf16_mfu: 53.80 | total time: 67.40m | eta: 43.4m step 03391/05568 (60.90%) | loss: 2.519776 | lrm: 0.62 | dt: 1190.78ms | tok/sec: 880,576 | bf16_mfu: 54.09 | total time: 67.42m | eta: 43.4m step 03392/05568 (60.92%) | loss: 2.516410 | lrm: 0.62 | dt: 1190.31ms | tok/sec: 880,926 | bf16_mfu: 54.11 | total time: 67.44m | eta: 43.4m step 03393/05568 (60.94%) | loss: 2.508952 | lrm: 0.62 | dt: 1192.75ms | tok/sec: 879,124 | bf16_mfu: 54.00 | total time: 67.46m | eta: 43.3m step 03394/05568 (60.96%) | loss: 2.505355 | lrm: 0.62 | dt: 1194.14ms | tok/sec: 878,102 | bf16_mfu: 53.94 | total time: 67.48m | eta: 43.3m step 03395/05568 (60.97%) | loss: 2.501547 | lrm: 0.62 | dt: 1194.05ms | tok/sec: 878,169 | bf16_mfu: 53.94 | total time: 67.50m | eta: 43.3m step 03396/05568 (60.99%) | loss: 2.500204 | lrm: 0.62 | dt: 1194.32ms | tok/sec: 877,966 | bf16_mfu: 53.93 | total time: 67.52m | eta: 43.3m step 03397/05568 (61.01%) | loss: 2.501639 | lrm: 0.62 | dt: 1194.06ms | tok/sec: 878,159 | bf16_mfu: 53.94 | total time: 67.53m | eta: 43.3m step 03398/05568 (61.03%) | loss: 2.510182 | lrm: 0.62 | dt: 1193.98ms | tok/sec: 878,218 | bf16_mfu: 53.95 | total time: 67.55m | eta: 43.2m step 03399/05568 (61.05%) | loss: 2.507179 | lrm: 0.62 | dt: 1197.03ms | tok/sec: 875,981 | bf16_mfu: 53.81 | total time: 67.57m | eta: 43.2m step 03400/05568 (61.06%) | loss: 2.496516 | lrm: 0.62 | dt: 1193.19ms | tok/sec: 878,800 | bf16_mfu: 53.98 | total time: 67.59m | eta: 43.2m step 03401/05568 (61.08%) | loss: 2.490004 | lrm: 0.62 | dt: 1190.88ms | tok/sec: 880,501 | bf16_mfu: 54.09 | total time: 67.61m | eta: 43.2m step 03402/05568 (61.10%) | loss: 2.499837 | lrm: 0.62 | dt: 1192.97ms | tok/sec: 878,960 | bf16_mfu: 53.99 | total time: 67.63m | eta: 43.2m step 03403/05568 (61.12%) | loss: 2.496627 | lrm: 0.62 | dt: 1196.38ms | tok/sec: 876,455 | bf16_mfu: 53.84 | total time: 67.65m | eta: 43.1m step 03404/05568 (61.14%) | loss: 2.506292 | lrm: 0.62 | dt: 1195.12ms | tok/sec: 877,384 | bf16_mfu: 53.90 | total time: 67.67m | eta: 43.1m step 03405/05568 (61.15%) | loss: 2.495715 | lrm: 0.62 | dt: 1190.24ms | tok/sec: 880,981 | bf16_mfu: 54.12 | total time: 67.69m | eta: 43.1m step 03406/05568 (61.17%) | loss: 2.494503 | lrm: 0.62 | dt: 1192.83ms | tok/sec: 879,062 | bf16_mfu: 54.00 | total time: 67.71m | eta: 43.1m step 03407/05568 (61.19%) | loss: 2.500065 | lrm: 0.62 | dt: 1191.54ms | tok/sec: 880,018 | bf16_mfu: 54.06 | total time: 67.73m | eta: 43.1m step 03408/05568 (61.21%) | loss: 2.498399 | lrm: 0.62 | dt: 1193.12ms | tok/sec: 878,854 | bf16_mfu: 53.99 | total time: 67.75m | eta: 43.0m step 03409/05568 (61.22%) | loss: 2.491845 | lrm: 0.62 | dt: 1190.61ms | tok/sec: 880,707 | bf16_mfu: 54.10 | total time: 67.77m | eta: 43.0m step 03410/05568 (61.24%) | loss: 2.503621 | lrm: 0.62 | dt: 1190.92ms | tok/sec: 880,476 | bf16_mfu: 54.09 | total time: 67.79m | eta: 43.0m step 03411/05568 (61.26%) | loss: 2.494471 | lrm: 0.62 | dt: 1194.47ms | tok/sec: 877,856 | bf16_mfu: 53.92 | total time: 67.81m | eta: 43.0m step 03412/05568 (61.28%) | loss: 2.495429 | lrm: 0.62 | dt: 1192.82ms | tok/sec: 879,074 | bf16_mfu: 54.00 | total time: 67.83m | eta: 43.0m step 03413/05568 (61.30%) | loss: 2.485599 | lrm: 0.62 | dt: 1191.61ms | tok/sec: 879,967 | bf16_mfu: 54.05 | total time: 67.85m | eta: 42.9m step 03414/05568 (61.31%) | loss: 2.477752 | lrm: 0.62 | dt: 1192.73ms | tok/sec: 879,141 | bf16_mfu: 54.00 | total time: 67.87m | eta: 42.9m step 03415/05568 (61.33%) | loss: 2.470649 | lrm: 0.62 | dt: 1190.24ms | tok/sec: 880,981 | bf16_mfu: 54.12 | total time: 67.89m | eta: 42.9m step 03416/05568 (61.35%) | loss: 2.483771 | lrm: 0.61 | dt: 1194.34ms | tok/sec: 877,956 | bf16_mfu: 53.93 | total time: 67.91m | eta: 42.9m step 03417/05568 (61.37%) | loss: 2.488770 | lrm: 0.61 | dt: 1193.21ms | tok/sec: 878,786 | bf16_mfu: 53.98 | total time: 67.93m | eta: 42.9m step 03418/05568 (61.39%) | loss: 2.489931 | lrm: 0.61 | dt: 1190.59ms | tok/sec: 880,716 | bf16_mfu: 54.10 | total time: 67.95m | eta: 42.8m step 03419/05568 (61.40%) | loss: 2.492135 | lrm: 0.61 | dt: 1191.01ms | tok/sec: 880,408 | bf16_mfu: 54.08 | total time: 67.97m | eta: 42.8m step 03420/05568 (61.42%) | loss: 2.499141 | lrm: 0.61 | dt: 1192.43ms | tok/sec: 879,362 | bf16_mfu: 54.02 | total time: 67.99m | eta: 42.8m step 03421/05568 (61.44%) | loss: 2.505001 | lrm: 0.61 | dt: 1191.41ms | tok/sec: 880,113 | bf16_mfu: 54.06 | total time: 68.01m | eta: 42.8m step 03422/05568 (61.46%) | loss: 2.516621 | lrm: 0.61 | dt: 1197.30ms | tok/sec: 875,782 | bf16_mfu: 53.80 | total time: 68.03m | eta: 42.8m step 03423/05568 (61.48%) | loss: 2.510589 | lrm: 0.61 | dt: 1191.58ms | tok/sec: 879,986 | bf16_mfu: 54.06 | total time: 68.05m | eta: 42.7m step 03424/05568 (61.49%) | loss: 2.500204 | lrm: 0.61 | dt: 1189.94ms | tok/sec: 881,204 | bf16_mfu: 54.13 | total time: 68.07m | eta: 42.7m step 03425/05568 (61.51%) | loss: 2.501116 | lrm: 0.61 | dt: 1193.53ms | tok/sec: 878,553 | bf16_mfu: 53.97 | total time: 68.09m | eta: 42.7m step 03426/05568 (61.53%) | loss: 2.488397 | lrm: 0.61 | dt: 1190.43ms | tok/sec: 880,834 | bf16_mfu: 54.11 | total time: 68.11m | eta: 42.7m step 03427/05568 (61.55%) | loss: 2.489441 | lrm: 0.61 | dt: 1193.95ms | tok/sec: 878,240 | bf16_mfu: 53.95 | total time: 68.13m | eta: 42.7m step 03428/05568 (61.57%) | loss: 2.494873 | lrm: 0.61 | dt: 1197.10ms | tok/sec: 875,930 | bf16_mfu: 53.81 | total time: 68.15m | eta: 42.6m step 03429/05568 (61.58%) | loss: 2.501895 | lrm: 0.61 | dt: 1190.94ms | tok/sec: 880,457 | bf16_mfu: 54.08 | total time: 68.17m | eta: 42.6m step 03430/05568 (61.60%) | loss: 2.507275 | lrm: 0.61 | dt: 1192.71ms | tok/sec: 879,157 | bf16_mfu: 54.00 | total time: 68.19m | eta: 42.6m step 03431/05568 (61.62%) | loss: 2.511621 | lrm: 0.61 | dt: 1193.53ms | tok/sec: 878,550 | bf16_mfu: 53.97 | total time: 68.21m | eta: 42.6m step 03432/05568 (61.64%) | loss: 2.517817 | lrm: 0.61 | dt: 1188.43ms | tok/sec: 882,319 | bf16_mfu: 54.20 | total time: 68.23m | eta: 42.6m step 03433/05568 (61.66%) | loss: 2.510628 | lrm: 0.61 | dt: 1192.98ms | tok/sec: 878,953 | bf16_mfu: 53.99 | total time: 68.25m | eta: 42.5m step 03434/05568 (61.67%) | loss: 2.497756 | lrm: 0.61 | dt: 1192.34ms | tok/sec: 879,427 | bf16_mfu: 54.02 | total time: 68.27m | eta: 42.5m step 03435/05568 (61.69%) | loss: 2.504419 | lrm: 0.61 | dt: 1190.14ms | tok/sec: 881,055 | bf16_mfu: 54.12 | total time: 68.29m | eta: 42.5m step 03436/05568 (61.71%) | loss: 2.511330 | lrm: 0.61 | dt: 1187.31ms | tok/sec: 883,152 | bf16_mfu: 54.25 | total time: 68.31m | eta: 42.5m step 03437/05568 (61.73%) | loss: 2.502617 | lrm: 0.61 | dt: 1189.97ms | tok/sec: 881,178 | bf16_mfu: 54.13 | total time: 68.33m | eta: 42.5m step 03438/05568 (61.75%) | loss: 2.495819 | lrm: 0.61 | dt: 1192.62ms | tok/sec: 879,222 | bf16_mfu: 54.01 | total time: 68.35m | eta: 42.4m step 03439/05568 (61.76%) | loss: 2.495223 | lrm: 0.61 | dt: 1194.94ms | tok/sec: 877,515 | bf16_mfu: 53.90 | total time: 68.37m | eta: 42.4m step 03440/05568 (61.78%) | loss: 2.507885 | lrm: 0.61 | dt: 1191.79ms | tok/sec: 879,834 | bf16_mfu: 54.05 | total time: 68.39m | eta: 42.4m step 03441/05568 (61.80%) | loss: 2.505210 | lrm: 0.61 | dt: 1186.16ms | tok/sec: 884,010 | bf16_mfu: 54.30 | total time: 68.41m | eta: 42.4m step 03442/05568 (61.82%) | loss: 2.508370 | lrm: 0.61 | dt: 1196.12ms | tok/sec: 876,650 | bf16_mfu: 53.85 | total time: 68.43m | eta: 42.4m step 03443/05568 (61.84%) | loss: 2.510049 | lrm: 0.61 | dt: 1193.74ms | tok/sec: 878,397 | bf16_mfu: 53.96 | total time: 68.45m | eta: 42.3m step 03444/05568 (61.85%) | loss: 2.498508 | lrm: 0.61 | dt: 1189.06ms | tok/sec: 881,850 | bf16_mfu: 54.17 | total time: 68.47m | eta: 42.3m step 03445/05568 (61.87%) | loss: 2.499117 | lrm: 0.61 | dt: 1190.61ms | tok/sec: 880,704 | bf16_mfu: 54.10 | total time: 68.49m | eta: 42.3m step 03446/05568 (61.89%) | loss: 2.498330 | lrm: 0.61 | dt: 1189.46ms | tok/sec: 881,559 | bf16_mfu: 54.15 | total time: 68.51m | eta: 42.3m step 03447/05568 (61.91%) | loss: 2.498531 | lrm: 0.61 | dt: 1190.73ms | tok/sec: 880,616 | bf16_mfu: 54.09 | total time: 68.53m | eta: 42.3m step 03448/05568 (61.93%) | loss: 2.505812 | lrm: 0.61 | dt: 1194.55ms | tok/sec: 877,799 | bf16_mfu: 53.92 | total time: 68.55m | eta: 42.2m step 03449/05568 (61.94%) | loss: 2.506104 | lrm: 0.61 | dt: 1189.92ms | tok/sec: 881,214 | bf16_mfu: 54.13 | total time: 68.57m | eta: 42.2m step 03450/05568 (61.96%) | loss: 2.499701 | lrm: 0.61 | dt: 1186.96ms | tok/sec: 883,415 | bf16_mfu: 54.27 | total time: 68.59m | eta: 42.2m step 03451/05568 (61.98%) | loss: 2.507485 | lrm: 0.61 | dt: 1196.53ms | tok/sec: 876,343 | bf16_mfu: 53.83 | total time: 68.61m | eta: 42.2m step 03452/05568 (62.00%) | loss: 2.510626 | lrm: 0.61 | dt: 1196.13ms | tok/sec: 876,638 | bf16_mfu: 53.85 | total time: 68.63m | eta: 42.2m step 03453/05568 (62.02%) | loss: 2.507846 | lrm: 0.61 | dt: 1191.74ms | tok/sec: 879,869 | bf16_mfu: 54.05 | total time: 68.65m | eta: 42.1m step 03454/05568 (62.03%) | loss: 2.509820 | lrm: 0.60 | dt: 1191.32ms | tok/sec: 880,177 | bf16_mfu: 54.07 | total time: 68.67m | eta: 42.1m step 03455/05568 (62.05%) | loss: 2.516311 | lrm: 0.60 | dt: 1194.52ms | tok/sec: 877,825 | bf16_mfu: 53.92 | total time: 68.69m | eta: 42.1m step 03456/05568 (62.07%) | loss: 2.515311 | lrm: 0.60 | dt: 1195.52ms | tok/sec: 877,085 | bf16_mfu: 53.88 | total time: 68.71m | eta: 42.1m step 03457/05568 (62.09%) | loss: 2.521573 | lrm: 0.60 | dt: 1190.91ms | tok/sec: 880,483 | bf16_mfu: 54.09 | total time: 68.73m | eta: 42.1m step 03458/05568 (62.10%) | loss: 2.527446 | lrm: 0.60 | dt: 1192.88ms | tok/sec: 879,029 | bf16_mfu: 54.00 | total time: 68.75m | eta: 42.0m step 03459/05568 (62.12%) | loss: 2.518157 | lrm: 0.60 | dt: 1189.82ms | tok/sec: 881,287 | bf16_mfu: 54.14 | total time: 68.77m | eta: 42.0m step 03460/05568 (62.14%) | loss: 2.517426 | lrm: 0.60 | dt: 1190.14ms | tok/sec: 881,054 | bf16_mfu: 54.12 | total time: 68.79m | eta: 42.0m step 03461/05568 (62.16%) | loss: 2.520612 | lrm: 0.60 | dt: 1195.87ms | tok/sec: 876,834 | bf16_mfu: 53.86 | total time: 68.81m | eta: 42.0m step 03462/05568 (62.18%) | loss: 2.515510 | lrm: 0.60 | dt: 1192.84ms | tok/sec: 879,060 | bf16_mfu: 54.00 | total time: 68.83m | eta: 42.0m step 03463/05568 (62.19%) | loss: 2.509196 | lrm: 0.60 | dt: 1190.90ms | tok/sec: 880,490 | bf16_mfu: 54.09 | total time: 68.85m | eta: 41.9m step 03464/05568 (62.21%) | loss: 2.506558 | lrm: 0.60 | dt: 1189.65ms | tok/sec: 881,415 | bf16_mfu: 54.14 | total time: 68.87m | eta: 41.9m step 03465/05568 (62.23%) | loss: 2.510603 | lrm: 0.60 | dt: 1194.54ms | tok/sec: 877,806 | bf16_mfu: 53.92 | total time: 68.89m | eta: 41.9m step 03466/05568 (62.25%) | loss: 2.506430 | lrm: 0.60 | dt: 1195.09ms | tok/sec: 877,401 | bf16_mfu: 53.90 | total time: 68.91m | eta: 41.9m step 03467/05568 (62.27%) | loss: 2.505555 | lrm: 0.60 | dt: 1191.57ms | tok/sec: 879,998 | bf16_mfu: 54.06 | total time: 68.93m | eta: 41.9m step 03468/05568 (62.28%) | loss: 2.502376 | lrm: 0.60 | dt: 1194.06ms | tok/sec: 878,157 | bf16_mfu: 53.94 | total time: 68.95m | eta: 41.9m step 03469/05568 (62.30%) | loss: 2.507075 | lrm: 0.60 | dt: 1191.91ms | tok/sec: 879,746 | bf16_mfu: 54.04 | total time: 68.97m | eta: 41.8m step 03470/05568 (62.32%) | loss: 2.497919 | lrm: 0.60 | dt: 1190.35ms | tok/sec: 880,896 | bf16_mfu: 54.11 | total time: 68.99m | eta: 41.8m step 03471/05568 (62.34%) | loss: 2.493323 | lrm: 0.60 | dt: 1194.73ms | tok/sec: 877,670 | bf16_mfu: 53.91 | total time: 69.01m | eta: 41.8m step 03472/05568 (62.36%) | loss: 2.498213 | lrm: 0.60 | dt: 1190.33ms | tok/sec: 880,914 | bf16_mfu: 54.11 | total time: 69.03m | eta: 41.8m step 03473/05568 (62.37%) | loss: 2.497214 | lrm: 0.60 | dt: 1192.79ms | tok/sec: 879,094 | bf16_mfu: 54.00 | total time: 69.05m | eta: 41.8m step 03474/05568 (62.39%) | loss: 2.499403 | lrm: 0.60 | dt: 1195.01ms | tok/sec: 877,461 | bf16_mfu: 53.90 | total time: 69.07m | eta: 41.7m step 03475/05568 (62.41%) | loss: 2.485517 | lrm: 0.60 | dt: 1192.35ms | tok/sec: 879,422 | bf16_mfu: 54.02 | total time: 69.09m | eta: 41.7m step 03476/05568 (62.43%) | loss: 2.490529 | lrm: 0.60 | dt: 1194.24ms | tok/sec: 878,029 | bf16_mfu: 53.94 | total time: 69.10m | eta: 41.7m step 03477/05568 (62.45%) | loss: 2.486586 | lrm: 0.60 | dt: 1192.31ms | tok/sec: 879,452 | bf16_mfu: 54.02 | total time: 69.12m | eta: 41.7m step 03478/05568 (62.46%) | loss: 2.478452 | lrm: 0.60 | dt: 1196.61ms | tok/sec: 876,287 | bf16_mfu: 53.83 | total time: 69.14m | eta: 41.7m step 03479/05568 (62.48%) | loss: 2.479751 | lrm: 0.60 | dt: 1193.03ms | tok/sec: 878,919 | bf16_mfu: 53.99 | total time: 69.16m | eta: 41.6m step 03480/05568 (62.50%) | loss: 2.466412 | lrm: 0.60 | dt: 1188.83ms | tok/sec: 882,023 | bf16_mfu: 54.18 | total time: 69.18m | eta: 41.6m step 03481/05568 (62.52%) | loss: 2.472960 | lrm: 0.60 | dt: 1190.74ms | tok/sec: 880,607 | bf16_mfu: 54.09 | total time: 69.20m | eta: 41.6m step 03482/05568 (62.54%) | loss: 2.482609 | lrm: 0.60 | dt: 1191.95ms | tok/sec: 879,715 | bf16_mfu: 54.04 | total time: 69.22m | eta: 41.6m step 03483/05568 (62.55%) | loss: 2.481057 | lrm: 0.60 | dt: 1193.57ms | tok/sec: 878,521 | bf16_mfu: 53.97 | total time: 69.24m | eta: 41.6m step 03484/05568 (62.57%) | loss: 2.494127 | lrm: 0.60 | dt: 1191.75ms | tok/sec: 879,865 | bf16_mfu: 54.05 | total time: 69.26m | eta: 41.5m step 03485/05568 (62.59%) | loss: 2.499867 | lrm: 0.60 | dt: 1195.21ms | tok/sec: 877,316 | bf16_mfu: 53.89 | total time: 69.28m | eta: 41.5m step 03486/05568 (62.61%) | loss: 2.506140 | lrm: 0.60 | dt: 1187.83ms | tok/sec: 882,765 | bf16_mfu: 54.23 | total time: 69.30m | eta: 41.5m step 03487/05568 (62.63%) | loss: 2.498182 | lrm: 0.60 | dt: 1191.51ms | tok/sec: 880,041 | bf16_mfu: 54.06 | total time: 69.32m | eta: 41.5m step 03488/05568 (62.64%) | loss: 2.501697 | lrm: 0.60 | dt: 1188.94ms | tok/sec: 881,941 | bf16_mfu: 54.18 | total time: 69.34m | eta: 41.5m step 03489/05568 (62.66%) | loss: 2.497701 | lrm: 0.60 | dt: 1194.76ms | tok/sec: 877,643 | bf16_mfu: 53.91 | total time: 69.36m | eta: 41.4m step 03490/05568 (62.68%) | loss: 2.502194 | lrm: 0.60 | dt: 1192.69ms | tok/sec: 879,172 | bf16_mfu: 54.01 | total time: 69.38m | eta: 41.4m step 03491/05568 (62.70%) | loss: 2.507493 | lrm: 0.60 | dt: 1187.73ms | tok/sec: 882,841 | bf16_mfu: 54.23 | total time: 69.40m | eta: 41.4m step 03492/05568 (62.72%) | loss: 2.514971 | lrm: 0.59 | dt: 1192.52ms | tok/sec: 879,297 | bf16_mfu: 54.01 | total time: 69.42m | eta: 41.4m step 03493/05568 (62.73%) | loss: 2.505273 | lrm: 0.59 | dt: 1195.61ms | tok/sec: 877,024 | bf16_mfu: 53.87 | total time: 69.44m | eta: 41.4m step 03494/05568 (62.75%) | loss: 2.502271 | lrm: 0.59 | dt: 1187.17ms | tok/sec: 883,258 | bf16_mfu: 54.26 | total time: 69.46m | eta: 41.3m step 03495/05568 (62.77%) | loss: 2.516682 | lrm: 0.59 | dt: 1191.97ms | tok/sec: 879,699 | bf16_mfu: 54.04 | total time: 69.48m | eta: 41.3m step 03496/05568 (62.79%) | loss: 2.505203 | lrm: 0.59 | dt: 1192.36ms | tok/sec: 879,412 | bf16_mfu: 54.02 | total time: 69.50m | eta: 41.3m step 03497/05568 (62.81%) | loss: 2.506044 | lrm: 0.59 | dt: 1195.25ms | tok/sec: 877,283 | bf16_mfu: 53.89 | total time: 69.52m | eta: 41.3m step 03498/05568 (62.82%) | loss: 2.505816 | lrm: 0.59 | dt: 1188.76ms | tok/sec: 882,073 | bf16_mfu: 54.18 | total time: 69.54m | eta: 41.3m step 03499/05568 (62.84%) | loss: 2.502704 | lrm: 0.59 | dt: 1191.37ms | tok/sec: 880,142 | bf16_mfu: 54.07 | total time: 69.56m | eta: 41.2m step:3500/5568 val_bpb:0.759539 val_time:3.90s step 03500/05568 (62.86%) | loss: 2.511542 | lrm: 0.59 | dt: 1194.43ms | tok/sec: 877,889 | bf16_mfu: 53.93 | total time: 69.58m | eta: 41.2m step 03501/05568 (62.88%) | loss: 2.510670 | lrm: 0.59 | dt: 1190.28ms | tok/sec: 880,952 | bf16_mfu: 54.11 | total time: 69.60m | eta: 41.2m step 03502/05568 (62.90%) | loss: 2.510438 | lrm: 0.59 | dt: 1193.18ms | tok/sec: 878,805 | bf16_mfu: 53.98 | total time: 69.62m | eta: 41.2m step 03503/05568 (62.91%) | loss: 2.501979 | lrm: 0.59 | dt: 1195.23ms | tok/sec: 877,299 | bf16_mfu: 53.89 | total time: 69.64m | eta: 41.2m step 03504/05568 (62.93%) | loss: 2.505041 | lrm: 0.59 | dt: 1190.92ms | tok/sec: 880,477 | bf16_mfu: 54.09 | total time: 69.66m | eta: 41.1m step 03505/05568 (62.95%) | loss: 2.492303 | lrm: 0.59 | dt: 1193.40ms | tok/sec: 878,643 | bf16_mfu: 53.97 | total time: 69.68m | eta: 41.1m step 03506/05568 (62.97%) | loss: 2.483821 | lrm: 0.59 | dt: 1189.39ms | tok/sec: 881,611 | bf16_mfu: 54.16 | total time: 69.70m | eta: 41.1m step 03507/05568 (62.98%) | loss: 2.488137 | lrm: 0.59 | dt: 1198.45ms | tok/sec: 874,944 | bf16_mfu: 53.75 | total time: 69.72m | eta: 41.1m step 03508/05568 (63.00%) | loss: 2.474514 | lrm: 0.59 | dt: 1194.68ms | tok/sec: 877,703 | bf16_mfu: 53.92 | total time: 69.74m | eta: 41.1m step 03509/05568 (63.02%) | loss: 2.471679 | lrm: 0.59 | dt: 1192.24ms | tok/sec: 879,500 | bf16_mfu: 54.03 | total time: 69.76m | eta: 41.0m step 03510/05568 (63.04%) | loss: 2.471993 | lrm: 0.59 | dt: 1190.27ms | tok/sec: 880,956 | bf16_mfu: 54.12 | total time: 69.78m | eta: 41.0m step 03511/05568 (63.06%) | loss: 2.481100 | lrm: 0.59 | dt: 1192.50ms | tok/sec: 879,307 | bf16_mfu: 54.01 | total time: 69.80m | eta: 41.0m step 03512/05568 (63.07%) | loss: 2.485311 | lrm: 0.59 | dt: 1197.34ms | tok/sec: 875,754 | bf16_mfu: 53.80 | total time: 69.82m | eta: 41.0m step 03513/05568 (63.09%) | loss: 2.490380 | lrm: 0.59 | dt: 1194.49ms | tok/sec: 877,843 | bf16_mfu: 53.92 | total time: 69.84m | eta: 41.0m step 03514/05568 (63.11%) | loss: 2.488553 | lrm: 0.59 | dt: 1198.62ms | tok/sec: 874,819 | bf16_mfu: 53.74 | total time: 69.86m | eta: 40.9m step 03515/05568 (63.13%) | loss: 2.491125 | lrm: 0.59 | dt: 1195.32ms | tok/sec: 877,231 | bf16_mfu: 53.89 | total time: 69.88m | eta: 40.9m step 03516/05568 (63.15%) | loss: 2.500489 | lrm: 0.59 | dt: 1197.91ms | tok/sec: 875,335 | bf16_mfu: 53.77 | total time: 69.90m | eta: 40.9m step 03517/05568 (63.16%) | loss: 2.500029 | lrm: 0.59 | dt: 1195.63ms | tok/sec: 877,009 | bf16_mfu: 53.87 | total time: 69.92m | eta: 40.9m step 03518/05568 (63.18%) | loss: 2.497204 | lrm: 0.59 | dt: 1197.26ms | tok/sec: 875,813 | bf16_mfu: 53.80 | total time: 69.94m | eta: 40.9m step 03519/05568 (63.20%) | loss: 2.504923 | lrm: 0.59 | dt: 1199.48ms | tok/sec: 874,188 | bf16_mfu: 53.70 | total time: 69.96m | eta: 40.8m step 03520/05568 (63.22%) | loss: 2.500827 | lrm: 0.59 | dt: 1189.68ms | tok/sec: 881,392 | bf16_mfu: 54.14 | total time: 69.98m | eta: 40.8m step 03521/05568 (63.24%) | loss: 2.503097 | lrm: 0.59 | dt: 1193.22ms | tok/sec: 878,777 | bf16_mfu: 53.98 | total time: 70.00m | eta: 40.8m step 03522/05568 (63.25%) | loss: 2.505400 | lrm: 0.59 | dt: 1195.60ms | tok/sec: 877,025 | bf16_mfu: 53.87 | total time: 70.02m | eta: 40.8m step 03523/05568 (63.27%) | loss: 2.499157 | lrm: 0.59 | dt: 1192.49ms | tok/sec: 879,315 | bf16_mfu: 54.01 | total time: 70.04m | eta: 40.8m step 03524/05568 (63.29%) | loss: 2.511101 | lrm: 0.59 | dt: 1193.75ms | tok/sec: 878,387 | bf16_mfu: 53.96 | total time: 70.06m | eta: 40.7m step 03525/05568 (63.31%) | loss: 2.512771 | lrm: 0.59 | dt: 1194.03ms | tok/sec: 878,180 | bf16_mfu: 53.94 | total time: 70.08m | eta: 40.7m step 03526/05568 (63.33%) | loss: 2.509546 | lrm: 0.59 | dt: 1192.21ms | tok/sec: 879,519 | bf16_mfu: 54.03 | total time: 70.10m | eta: 40.7m step 03527/05568 (63.34%) | loss: 2.495500 | lrm: 0.59 | dt: 1191.48ms | tok/sec: 880,061 | bf16_mfu: 54.06 | total time: 70.12m | eta: 40.7m step 03528/05568 (63.36%) | loss: 2.512042 | lrm: 0.59 | dt: 1193.58ms | tok/sec: 878,516 | bf16_mfu: 53.97 | total time: 70.14m | eta: 40.7m step 03529/05568 (63.38%) | loss: 2.513067 | lrm: 0.59 | dt: 1193.55ms | tok/sec: 878,536 | bf16_mfu: 53.97 | total time: 70.16m | eta: 40.6m step 03530/05568 (63.40%) | loss: 2.509418 | lrm: 0.58 | dt: 1191.96ms | tok/sec: 879,705 | bf16_mfu: 54.04 | total time: 70.18m | eta: 40.6m step 03531/05568 (63.42%) | loss: 2.503243 | lrm: 0.58 | dt: 1195.74ms | tok/sec: 876,923 | bf16_mfu: 53.87 | total time: 70.20m | eta: 40.6m step 03532/05568 (63.43%) | loss: 2.494672 | lrm: 0.58 | dt: 1192.05ms | tok/sec: 879,642 | bf16_mfu: 54.03 | total time: 70.22m | eta: 40.6m step 03533/05568 (63.45%) | loss: 2.495540 | lrm: 0.58 | dt: 1196.45ms | tok/sec: 876,405 | bf16_mfu: 53.84 | total time: 70.24m | eta: 40.6m step 03534/05568 (63.47%) | loss: 2.496166 | lrm: 0.58 | dt: 1193.89ms | tok/sec: 878,282 | bf16_mfu: 53.95 | total time: 70.26m | eta: 40.5m step 03535/05568 (63.49%) | loss: 2.491705 | lrm: 0.58 | dt: 1192.85ms | tok/sec: 879,053 | bf16_mfu: 54.00 | total time: 70.28m | eta: 40.5m step 03536/05568 (63.51%) | loss: 2.492690 | lrm: 0.58 | dt: 1192.39ms | tok/sec: 879,391 | bf16_mfu: 54.02 | total time: 70.30m | eta: 40.5m step 03537/05568 (63.52%) | loss: 2.490017 | lrm: 0.58 | dt: 1192.91ms | tok/sec: 879,007 | bf16_mfu: 54.00 | total time: 70.32m | eta: 40.5m step 03538/05568 (63.54%) | loss: 2.492914 | lrm: 0.58 | dt: 1195.70ms | tok/sec: 876,957 | bf16_mfu: 53.87 | total time: 70.34m | eta: 40.5m step 03539/05568 (63.56%) | loss: 2.498592 | lrm: 0.58 | dt: 1193.96ms | tok/sec: 878,237 | bf16_mfu: 53.95 | total time: 70.36m | eta: 40.4m step 03540/05568 (63.58%) | loss: 2.499663 | lrm: 0.58 | dt: 1195.81ms | tok/sec: 876,878 | bf16_mfu: 53.86 | total time: 70.38m | eta: 40.4m step 03541/05568 (63.60%) | loss: 2.492522 | lrm: 0.58 | dt: 1191.94ms | tok/sec: 879,724 | bf16_mfu: 54.04 | total time: 70.40m | eta: 40.4m step 03542/05568 (63.61%) | loss: 2.500249 | lrm: 0.58 | dt: 1194.33ms | tok/sec: 877,958 | bf16_mfu: 53.93 | total time: 70.42m | eta: 40.4m step 03543/05568 (63.63%) | loss: 2.497300 | lrm: 0.58 | dt: 1194.22ms | tok/sec: 878,040 | bf16_mfu: 53.94 | total time: 70.44m | eta: 40.4m step 03544/05568 (63.65%) | loss: 2.493225 | lrm: 0.58 | dt: 1195.40ms | tok/sec: 877,178 | bf16_mfu: 53.88 | total time: 70.46m | eta: 40.3m step 03545/05568 (63.67%) | loss: 2.497460 | lrm: 0.58 | dt: 1197.79ms | tok/sec: 875,422 | bf16_mfu: 53.78 | total time: 70.48m | eta: 40.3m step 03546/05568 (63.69%) | loss: 2.501899 | lrm: 0.58 | dt: 1194.80ms | tok/sec: 877,613 | bf16_mfu: 53.91 | total time: 70.50m | eta: 40.3m step 03547/05568 (63.70%) | loss: 2.509301 | lrm: 0.58 | dt: 1197.43ms | tok/sec: 875,686 | bf16_mfu: 53.79 | total time: 70.52m | eta: 40.3m step 03548/05568 (63.72%) | loss: 2.503961 | lrm: 0.58 | dt: 1192.48ms | tok/sec: 879,327 | bf16_mfu: 54.02 | total time: 70.54m | eta: 40.3m step 03549/05568 (63.74%) | loss: 2.504502 | lrm: 0.58 | dt: 1194.38ms | tok/sec: 877,927 | bf16_mfu: 53.93 | total time: 70.56m | eta: 40.2m step 03550/05568 (63.76%) | loss: 2.496581 | lrm: 0.58 | dt: 1193.11ms | tok/sec: 878,862 | bf16_mfu: 53.99 | total time: 70.58m | eta: 40.2m step 03551/05568 (63.78%) | loss: 2.504317 | lrm: 0.58 | dt: 1195.62ms | tok/sec: 877,011 | bf16_mfu: 53.87 | total time: 70.60m | eta: 40.2m step 03552/05568 (63.79%) | loss: 2.491099 | lrm: 0.58 | dt: 1193.77ms | tok/sec: 878,370 | bf16_mfu: 53.96 | total time: 70.62m | eta: 40.2m step 03553/05568 (63.81%) | loss: 2.506005 | lrm: 0.58 | dt: 1194.90ms | tok/sec: 877,543 | bf16_mfu: 53.91 | total time: 70.64m | eta: 40.2m step 03554/05568 (63.83%) | loss: 2.502871 | lrm: 0.58 | dt: 1191.90ms | tok/sec: 879,752 | bf16_mfu: 54.04 | total time: 70.66m | eta: 40.1m step 03555/05568 (63.85%) | loss: 2.501345 | lrm: 0.58 | dt: 1195.37ms | tok/sec: 877,196 | bf16_mfu: 53.88 | total time: 70.68m | eta: 40.1m step 03556/05568 (63.86%) | loss: 2.503186 | lrm: 0.58 | dt: 1195.99ms | tok/sec: 876,739 | bf16_mfu: 53.86 | total time: 70.70m | eta: 40.1m step 03557/05568 (63.88%) | loss: 2.498565 | lrm: 0.58 | dt: 1192.87ms | tok/sec: 879,035 | bf16_mfu: 54.00 | total time: 70.72m | eta: 40.1m step 03558/05568 (63.90%) | loss: 2.492039 | lrm: 0.58 | dt: 1203.03ms | tok/sec: 871,611 | bf16_mfu: 53.54 | total time: 70.74m | eta: 40.1m step 03559/05568 (63.92%) | loss: 2.506185 | lrm: 0.58 | dt: 1193.74ms | tok/sec: 878,393 | bf16_mfu: 53.96 | total time: 70.76m | eta: 40.0m step 03560/05568 (63.94%) | loss: 2.503693 | lrm: 0.58 | dt: 1192.23ms | tok/sec: 879,510 | bf16_mfu: 54.03 | total time: 70.78m | eta: 40.0m step 03561/05568 (63.95%) | loss: 2.490208 | lrm: 0.58 | dt: 1195.79ms | tok/sec: 876,888 | bf16_mfu: 53.87 | total time: 70.80m | eta: 40.0m step 03562/05568 (63.97%) | loss: 2.476272 | lrm: 0.58 | dt: 1192.12ms | tok/sec: 879,591 | bf16_mfu: 54.03 | total time: 70.82m | eta: 40.0m step 03563/05568 (63.99%) | loss: 2.469851 | lrm: 0.58 | dt: 1193.41ms | tok/sec: 878,635 | bf16_mfu: 53.97 | total time: 70.84m | eta: 40.0m step 03564/05568 (64.01%) | loss: 2.473676 | lrm: 0.58 | dt: 1194.46ms | tok/sec: 877,864 | bf16_mfu: 53.93 | total time: 70.86m | eta: 39.9m step 03565/05568 (64.03%) | loss: 2.471298 | lrm: 0.58 | dt: 1196.08ms | tok/sec: 876,677 | bf16_mfu: 53.85 | total time: 70.88m | eta: 39.9m step 03566/05568 (64.04%) | loss: 2.476072 | lrm: 0.58 | dt: 1193.98ms | tok/sec: 878,215 | bf16_mfu: 53.95 | total time: 70.90m | eta: 39.9m step 03567/05568 (64.06%) | loss: 2.475705 | lrm: 0.58 | dt: 1195.39ms | tok/sec: 877,184 | bf16_mfu: 53.88 | total time: 70.92m | eta: 39.9m step 03568/05568 (64.08%) | loss: 2.462634 | lrm: 0.58 | dt: 1193.13ms | tok/sec: 878,846 | bf16_mfu: 53.99 | total time: 70.94m | eta: 39.9m step 03569/05568 (64.10%) | loss: 2.462740 | lrm: 0.57 | dt: 1193.12ms | tok/sec: 878,849 | bf16_mfu: 53.99 | total time: 70.96m | eta: 39.8m step 03570/05568 (64.12%) | loss: 2.470000 | lrm: 0.57 | dt: 1193.28ms | tok/sec: 878,732 | bf16_mfu: 53.98 | total time: 70.98m | eta: 39.8m step 03571/05568 (64.13%) | loss: 2.479047 | lrm: 0.57 | dt: 1191.18ms | tok/sec: 880,281 | bf16_mfu: 54.07 | total time: 70.99m | eta: 39.8m step 03572/05568 (64.15%) | loss: 2.479812 | lrm: 0.57 | dt: 1194.43ms | tok/sec: 877,891 | bf16_mfu: 53.93 | total time: 71.01m | eta: 39.8m step 03573/05568 (64.17%) | loss: 2.484206 | lrm: 0.57 | dt: 1191.97ms | tok/sec: 879,701 | bf16_mfu: 54.04 | total time: 71.03m | eta: 39.8m step 03574/05568 (64.19%) | loss: 2.474504 | lrm: 0.57 | dt: 1196.22ms | tok/sec: 876,571 | bf16_mfu: 53.85 | total time: 71.05m | eta: 39.7m step 03575/05568 (64.21%) | loss: 2.464736 | lrm: 0.57 | dt: 1192.67ms | tok/sec: 879,181 | bf16_mfu: 54.01 | total time: 71.07m | eta: 39.7m step 03576/05568 (64.22%) | loss: 2.454532 | lrm: 0.57 | dt: 1191.92ms | tok/sec: 879,740 | bf16_mfu: 54.04 | total time: 71.09m | eta: 39.7m step 03577/05568 (64.24%) | loss: 2.467617 | lrm: 0.57 | dt: 1196.53ms | tok/sec: 876,350 | bf16_mfu: 53.83 | total time: 71.11m | eta: 39.7m step 03578/05568 (64.26%) | loss: 2.470341 | lrm: 0.57 | dt: 1190.13ms | tok/sec: 881,056 | bf16_mfu: 54.12 | total time: 71.13m | eta: 39.7m step 03579/05568 (64.28%) | loss: 2.462585 | lrm: 0.57 | dt: 1196.04ms | tok/sec: 876,708 | bf16_mfu: 53.85 | total time: 71.15m | eta: 39.6m step 03580/05568 (64.30%) | loss: 2.460326 | lrm: 0.57 | dt: 1192.56ms | tok/sec: 879,264 | bf16_mfu: 54.01 | total time: 71.17m | eta: 39.6m step 03581/05568 (64.31%) | loss: 2.473130 | lrm: 0.57 | dt: 1194.53ms | tok/sec: 877,813 | bf16_mfu: 53.92 | total time: 71.19m | eta: 39.6m step 03582/05568 (64.33%) | loss: 2.488225 | lrm: 0.57 | dt: 1195.48ms | tok/sec: 877,114 | bf16_mfu: 53.88 | total time: 71.21m | eta: 39.6m step 03583/05568 (64.35%) | loss: 2.493728 | lrm: 0.57 | dt: 1195.08ms | tok/sec: 877,411 | bf16_mfu: 53.90 | total time: 71.23m | eta: 39.6m step 03584/05568 (64.37%) | loss: 2.507899 | lrm: 0.57 | dt: 1193.30ms | tok/sec: 878,718 | bf16_mfu: 53.98 | total time: 71.25m | eta: 39.5m step 03585/05568 (64.39%) | loss: 2.491606 | lrm: 0.57 | dt: 1194.81ms | tok/sec: 877,605 | bf16_mfu: 53.91 | total time: 71.27m | eta: 39.5m step 03586/05568 (64.40%) | loss: 2.488976 | lrm: 0.57 | dt: 1192.66ms | tok/sec: 879,187 | bf16_mfu: 54.01 | total time: 71.29m | eta: 39.5m step 03587/05568 (64.42%) | loss: 2.495674 | lrm: 0.57 | dt: 1196.00ms | tok/sec: 876,735 | bf16_mfu: 53.86 | total time: 71.31m | eta: 39.5m step 03588/05568 (64.44%) | loss: 2.491223 | lrm: 0.57 | dt: 1191.19ms | tok/sec: 880,279 | bf16_mfu: 54.07 | total time: 71.33m | eta: 39.5m step 03589/05568 (64.46%) | loss: 2.490847 | lrm: 0.57 | dt: 1194.88ms | tok/sec: 877,556 | bf16_mfu: 53.91 | total time: 71.35m | eta: 39.4m step 03590/05568 (64.48%) | loss: 2.490131 | lrm: 0.57 | dt: 1194.72ms | tok/sec: 877,672 | bf16_mfu: 53.91 | total time: 71.37m | eta: 39.4m step 03591/05568 (64.49%) | loss: 2.486231 | lrm: 0.57 | dt: 1200.74ms | tok/sec: 873,274 | bf16_mfu: 53.64 | total time: 71.39m | eta: 39.4m step 03592/05568 (64.51%) | loss: 2.490306 | lrm: 0.57 | dt: 1197.95ms | tok/sec: 875,309 | bf16_mfu: 53.77 | total time: 71.41m | eta: 39.4m step 03593/05568 (64.53%) | loss: 2.497530 | lrm: 0.57 | dt: 1197.55ms | tok/sec: 875,599 | bf16_mfu: 53.79 | total time: 71.43m | eta: 39.4m step 03594/05568 (64.55%) | loss: 2.495206 | lrm: 0.57 | dt: 1190.11ms | tok/sec: 881,075 | bf16_mfu: 54.12 | total time: 71.45m | eta: 39.3m step 03595/05568 (64.57%) | loss: 2.494642 | lrm: 0.57 | dt: 1191.22ms | tok/sec: 880,254 | bf16_mfu: 54.07 | total time: 71.47m | eta: 39.3m step 03596/05568 (64.58%) | loss: 2.491876 | lrm: 0.57 | dt: 1191.50ms | tok/sec: 880,046 | bf16_mfu: 54.06 | total time: 71.49m | eta: 39.3m step 03597/05568 (64.60%) | loss: 2.486505 | lrm: 0.57 | dt: 1191.21ms | tok/sec: 880,259 | bf16_mfu: 54.07 | total time: 71.51m | eta: 39.3m step 03598/05568 (64.62%) | loss: 2.484124 | lrm: 0.57 | dt: 1196.13ms | tok/sec: 876,644 | bf16_mfu: 53.85 | total time: 71.53m | eta: 39.3m step 03599/05568 (64.64%) | loss: 2.483214 | lrm: 0.57 | dt: 1193.54ms | tok/sec: 878,545 | bf16_mfu: 53.97 | total time: 71.55m | eta: 39.2m step 03600/05568 (64.66%) | loss: 2.480480 | lrm: 0.57 | dt: 1195.55ms | tok/sec: 877,064 | bf16_mfu: 53.88 | total time: 71.57m | eta: 39.2m step 03601/05568 (64.67%) | loss: 2.486400 | lrm: 0.57 | dt: 1197.94ms | tok/sec: 875,318 | bf16_mfu: 53.77 | total time: 71.59m | eta: 39.2m step 03602/05568 (64.69%) | loss: 2.491692 | lrm: 0.57 | dt: 1189.21ms | tok/sec: 881,744 | bf16_mfu: 54.16 | total time: 71.61m | eta: 39.2m step 03603/05568 (64.71%) | loss: 2.488247 | lrm: 0.57 | dt: 1191.75ms | tok/sec: 879,863 | bf16_mfu: 54.05 | total time: 71.63m | eta: 39.2m step 03604/05568 (64.73%) | loss: 2.493312 | lrm: 0.57 | dt: 1191.05ms | tok/sec: 880,380 | bf16_mfu: 54.08 | total time: 71.65m | eta: 39.1m step 03605/05568 (64.74%) | loss: 2.500483 | lrm: 0.57 | dt: 1196.46ms | tok/sec: 876,398 | bf16_mfu: 53.84 | total time: 71.67m | eta: 39.1m step 03606/05568 (64.76%) | loss: 2.499550 | lrm: 0.57 | dt: 1194.78ms | tok/sec: 877,634 | bf16_mfu: 53.91 | total time: 71.69m | eta: 39.1m step 03607/05568 (64.78%) | loss: 2.502675 | lrm: 0.56 | dt: 1191.22ms | tok/sec: 880,251 | bf16_mfu: 54.07 | total time: 71.71m | eta: 39.1m step 03608/05568 (64.80%) | loss: 2.503908 | lrm: 0.56 | dt: 1194.87ms | tok/sec: 877,563 | bf16_mfu: 53.91 | total time: 71.73m | eta: 39.1m step 03609/05568 (64.82%) | loss: 2.502316 | lrm: 0.56 | dt: 1195.28ms | tok/sec: 877,260 | bf16_mfu: 53.89 | total time: 71.75m | eta: 39.0m step 03610/05568 (64.83%) | loss: 2.504926 | lrm: 0.56 | dt: 1192.75ms | tok/sec: 879,124 | bf16_mfu: 54.00 | total time: 71.77m | eta: 39.0m step 03611/05568 (64.85%) | loss: 2.500740 | lrm: 0.56 | dt: 1193.27ms | tok/sec: 878,742 | bf16_mfu: 53.98 | total time: 71.79m | eta: 39.0m step 03612/05568 (64.87%) | loss: 2.495754 | lrm: 0.56 | dt: 1196.04ms | tok/sec: 876,706 | bf16_mfu: 53.85 | total time: 71.81m | eta: 39.0m step 03613/05568 (64.89%) | loss: 2.497169 | lrm: 0.56 | dt: 1191.28ms | tok/sec: 880,212 | bf16_mfu: 54.07 | total time: 71.83m | eta: 39.0m step 03614/05568 (64.91%) | loss: 2.496809 | lrm: 0.56 | dt: 1191.88ms | tok/sec: 879,769 | bf16_mfu: 54.04 | total time: 71.85m | eta: 38.9m step 03615/05568 (64.92%) | loss: 2.486849 | lrm: 0.56 | dt: 1196.99ms | tok/sec: 876,010 | bf16_mfu: 53.81 | total time: 71.87m | eta: 38.9m step 03616/05568 (64.94%) | loss: 2.492107 | lrm: 0.56 | dt: 1193.21ms | tok/sec: 878,782 | bf16_mfu: 53.98 | total time: 71.89m | eta: 38.9m step 03617/05568 (64.96%) | loss: 2.493486 | lrm: 0.56 | dt: 1190.34ms | tok/sec: 880,902 | bf16_mfu: 54.11 | total time: 71.91m | eta: 38.9m step 03618/05568 (64.98%) | loss: 2.496819 | lrm: 0.56 | dt: 1196.87ms | tok/sec: 876,101 | bf16_mfu: 53.82 | total time: 71.93m | eta: 38.9m step 03619/05568 (65.00%) | loss: 2.505309 | lrm: 0.56 | dt: 1193.47ms | tok/sec: 878,597 | bf16_mfu: 53.97 | total time: 71.95m | eta: 38.8m step 03620/05568 (65.01%) | loss: 2.513682 | lrm: 0.56 | dt: 1195.47ms | tok/sec: 877,124 | bf16_mfu: 53.88 | total time: 71.97m | eta: 38.8m step 03621/05568 (65.03%) | loss: 2.509886 | lrm: 0.56 | dt: 1192.47ms | tok/sec: 879,333 | bf16_mfu: 54.02 | total time: 71.99m | eta: 38.8m step 03622/05568 (65.05%) | loss: 2.499516 | lrm: 0.56 | dt: 1196.81ms | tok/sec: 876,139 | bf16_mfu: 53.82 | total time: 72.01m | eta: 38.8m step 03623/05568 (65.07%) | loss: 2.501085 | lrm: 0.56 | dt: 1189.02ms | tok/sec: 881,880 | bf16_mfu: 54.17 | total time: 72.03m | eta: 38.8m step 03624/05568 (65.09%) | loss: 2.492338 | lrm: 0.56 | dt: 1190.56ms | tok/sec: 880,741 | bf16_mfu: 54.10 | total time: 72.05m | eta: 38.7m step 03625/05568 (65.10%) | loss: 2.494110 | lrm: 0.56 | dt: 1194.17ms | tok/sec: 878,076 | bf16_mfu: 53.94 | total time: 72.07m | eta: 38.7m step 03626/05568 (65.12%) | loss: 2.488078 | lrm: 0.56 | dt: 1195.04ms | tok/sec: 877,438 | bf16_mfu: 53.90 | total time: 72.09m | eta: 38.7m step 03627/05568 (65.14%) | loss: 2.484510 | lrm: 0.56 | dt: 1196.27ms | tok/sec: 876,538 | bf16_mfu: 53.84 | total time: 72.11m | eta: 38.7m step 03628/05568 (65.16%) | loss: 2.483491 | lrm: 0.56 | dt: 1193.81ms | tok/sec: 878,341 | bf16_mfu: 53.95 | total time: 72.13m | eta: 38.7m step 03629/05568 (65.18%) | loss: 2.493634 | lrm: 0.56 | dt: 1194.66ms | tok/sec: 877,720 | bf16_mfu: 53.92 | total time: 72.15m | eta: 38.6m step 03630/05568 (65.19%) | loss: 2.500089 | lrm: 0.56 | dt: 1191.36ms | tok/sec: 880,150 | bf16_mfu: 54.07 | total time: 72.17m | eta: 38.6m step 03631/05568 (65.21%) | loss: 2.500407 | lrm: 0.56 | dt: 1195.06ms | tok/sec: 877,422 | bf16_mfu: 53.90 | total time: 72.19m | eta: 38.6m step 03632/05568 (65.23%) | loss: 2.508851 | lrm: 0.56 | dt: 1194.54ms | tok/sec: 877,805 | bf16_mfu: 53.92 | total time: 72.21m | eta: 38.6m step 03633/05568 (65.25%) | loss: 2.504917 | lrm: 0.56 | dt: 1193.53ms | tok/sec: 878,547 | bf16_mfu: 53.97 | total time: 72.23m | eta: 38.6m step 03634/05568 (65.27%) | loss: 2.503097 | lrm: 0.56 | dt: 1189.47ms | tok/sec: 881,550 | bf16_mfu: 54.15 | total time: 72.25m | eta: 38.5m step 03635/05568 (65.28%) | loss: 2.504705 | lrm: 0.56 | dt: 1195.40ms | tok/sec: 877,176 | bf16_mfu: 53.88 | total time: 72.27m | eta: 38.5m step 03636/05568 (65.30%) | loss: 2.508727 | lrm: 0.56 | dt: 1194.65ms | tok/sec: 877,726 | bf16_mfu: 53.92 | total time: 72.29m | eta: 38.5m step 03637/05568 (65.32%) | loss: 2.515313 | lrm: 0.56 | dt: 1194.12ms | tok/sec: 878,115 | bf16_mfu: 53.94 | total time: 72.31m | eta: 38.5m step 03638/05568 (65.34%) | loss: 2.508690 | lrm: 0.56 | dt: 1193.05ms | tok/sec: 878,905 | bf16_mfu: 53.99 | total time: 72.33m | eta: 38.5m step 03639/05568 (65.36%) | loss: 2.503971 | lrm: 0.56 | dt: 1190.41ms | tok/sec: 880,850 | bf16_mfu: 54.11 | total time: 72.35m | eta: 38.4m step 03640/05568 (65.37%) | loss: 2.503640 | lrm: 0.56 | dt: 1197.95ms | tok/sec: 875,307 | bf16_mfu: 53.77 | total time: 72.37m | eta: 38.4m step 03641/05568 (65.39%) | loss: 2.507772 | lrm: 0.56 | dt: 1194.37ms | tok/sec: 877,930 | bf16_mfu: 53.93 | total time: 72.39m | eta: 38.4m step 03642/05568 (65.41%) | loss: 2.505789 | lrm: 0.56 | dt: 1191.71ms | tok/sec: 879,888 | bf16_mfu: 54.05 | total time: 72.41m | eta: 38.4m step 03643/05568 (65.43%) | loss: 2.504892 | lrm: 0.56 | dt: 1192.81ms | tok/sec: 879,080 | bf16_mfu: 54.00 | total time: 72.43m | eta: 38.4m step 03644/05568 (65.45%) | loss: 2.504285 | lrm: 0.56 | dt: 1198.45ms | tok/sec: 874,941 | bf16_mfu: 53.75 | total time: 72.45m | eta: 38.3m step 03645/05568 (65.46%) | loss: 2.495776 | lrm: 0.55 | dt: 1193.53ms | tok/sec: 878,551 | bf16_mfu: 53.97 | total time: 72.47m | eta: 38.3m step 03646/05568 (65.48%) | loss: 2.495814 | lrm: 0.55 | dt: 1189.26ms | tok/sec: 881,704 | bf16_mfu: 54.16 | total time: 72.49m | eta: 38.3m step 03647/05568 (65.50%) | loss: 2.495204 | lrm: 0.55 | dt: 1192.57ms | tok/sec: 879,254 | bf16_mfu: 54.01 | total time: 72.51m | eta: 38.3m step 03648/05568 (65.52%) | loss: 2.491025 | lrm: 0.55 | dt: 1195.75ms | tok/sec: 876,922 | bf16_mfu: 53.87 | total time: 72.53m | eta: 38.3m step 03649/05568 (65.54%) | loss: 2.497609 | lrm: 0.55 | dt: 1198.66ms | tok/sec: 874,788 | bf16_mfu: 53.74 | total time: 72.55m | eta: 38.2m step 03650/05568 (65.55%) | loss: 2.488809 | lrm: 0.55 | dt: 1194.30ms | tok/sec: 877,986 | bf16_mfu: 53.93 | total time: 72.57m | eta: 38.2m step 03651/05568 (65.57%) | loss: 2.497804 | lrm: 0.55 | dt: 1194.01ms | tok/sec: 878,198 | bf16_mfu: 53.95 | total time: 72.59m | eta: 38.2m step 03652/05568 (65.59%) | loss: 2.504518 | lrm: 0.55 | dt: 1191.85ms | tok/sec: 879,785 | bf16_mfu: 54.04 | total time: 72.61m | eta: 38.2m step 03653/05568 (65.61%) | loss: 2.515637 | lrm: 0.55 | dt: 1191.70ms | tok/sec: 879,898 | bf16_mfu: 54.05 | total time: 72.63m | eta: 38.2m step 03654/05568 (65.62%) | loss: 2.522066 | lrm: 0.55 | dt: 1193.89ms | tok/sec: 878,285 | bf16_mfu: 53.95 | total time: 72.65m | eta: 38.1m step 03655/05568 (65.64%) | loss: 2.515474 | lrm: 0.55 | dt: 1199.03ms | tok/sec: 874,518 | bf16_mfu: 53.72 | total time: 72.67m | eta: 38.1m step 03656/05568 (65.66%) | loss: 2.516468 | lrm: 0.55 | dt: 1195.44ms | tok/sec: 877,146 | bf16_mfu: 53.88 | total time: 72.69m | eta: 38.1m step 03657/05568 (65.68%) | loss: 2.512416 | lrm: 0.55 | dt: 1191.77ms | tok/sec: 879,845 | bf16_mfu: 54.05 | total time: 72.71m | eta: 38.1m step 03658/05568 (65.70%) | loss: 2.515976 | lrm: 0.55 | dt: 1198.01ms | tok/sec: 875,262 | bf16_mfu: 53.77 | total time: 72.73m | eta: 38.1m step 03659/05568 (65.71%) | loss: 2.522050 | lrm: 0.55 | dt: 1194.05ms | tok/sec: 878,167 | bf16_mfu: 53.94 | total time: 72.75m | eta: 38.0m step 03660/05568 (65.73%) | loss: 2.516314 | lrm: 0.55 | dt: 1196.34ms | tok/sec: 876,489 | bf16_mfu: 53.84 | total time: 72.77m | eta: 38.0m step 03661/05568 (65.75%) | loss: 2.504501 | lrm: 0.55 | dt: 1193.66ms | tok/sec: 878,457 | bf16_mfu: 53.96 | total time: 72.79m | eta: 38.0m step 03662/05568 (65.77%) | loss: 2.492663 | lrm: 0.55 | dt: 1193.22ms | tok/sec: 878,781 | bf16_mfu: 53.98 | total time: 72.81m | eta: 38.0m step 03663/05568 (65.79%) | loss: 2.482727 | lrm: 0.55 | dt: 1194.80ms | tok/sec: 877,616 | bf16_mfu: 53.91 | total time: 72.83m | eta: 38.0m step 03664/05568 (65.80%) | loss: 2.490462 | lrm: 0.55 | dt: 1190.50ms | tok/sec: 880,787 | bf16_mfu: 54.10 | total time: 72.85m | eta: 37.9m step 03665/05568 (65.82%) | loss: 2.497747 | lrm: 0.55 | dt: 1195.98ms | tok/sec: 876,749 | bf16_mfu: 53.86 | total time: 72.87m | eta: 37.9m step 03666/05568 (65.84%) | loss: 2.495951 | lrm: 0.55 | dt: 1194.92ms | tok/sec: 877,527 | bf16_mfu: 53.90 | total time: 72.89m | eta: 37.9m step 03667/05568 (65.86%) | loss: 2.497528 | lrm: 0.55 | dt: 1192.53ms | tok/sec: 879,284 | bf16_mfu: 54.01 | total time: 72.91m | eta: 37.9m step 03668/05568 (65.88%) | loss: 2.503527 | lrm: 0.55 | dt: 1195.56ms | tok/sec: 877,054 | bf16_mfu: 53.88 | total time: 72.93m | eta: 37.9m step 03669/05568 (65.89%) | loss: 2.500794 | lrm: 0.55 | dt: 1199.74ms | tok/sec: 874,002 | bf16_mfu: 53.69 | total time: 72.95m | eta: 37.8m step 03670/05568 (65.91%) | loss: 2.502358 | lrm: 0.55 | dt: 1191.33ms | tok/sec: 880,169 | bf16_mfu: 54.07 | total time: 72.96m | eta: 37.8m step 03671/05568 (65.93%) | loss: 2.495469 | lrm: 0.55 | dt: 1193.11ms | tok/sec: 878,860 | bf16_mfu: 53.99 | total time: 72.98m | eta: 37.8m step 03672/05568 (65.95%) | loss: 2.493676 | lrm: 0.55 | dt: 1192.94ms | tok/sec: 878,985 | bf16_mfu: 53.99 | total time: 73.00m | eta: 37.8m step 03673/05568 (65.97%) | loss: 2.488785 | lrm: 0.55 | dt: 1193.64ms | tok/sec: 878,471 | bf16_mfu: 53.96 | total time: 73.02m | eta: 37.8m step 03674/05568 (65.98%) | loss: 2.486885 | lrm: 0.55 | dt: 1196.74ms | tok/sec: 876,195 | bf16_mfu: 53.82 | total time: 73.04m | eta: 37.7m step 03675/05568 (66.00%) | loss: 2.487202 | lrm: 0.55 | dt: 1191.26ms | tok/sec: 880,223 | bf16_mfu: 54.07 | total time: 73.06m | eta: 37.7m step 03676/05568 (66.02%) | loss: 2.490468 | lrm: 0.55 | dt: 1194.61ms | tok/sec: 877,759 | bf16_mfu: 53.92 | total time: 73.08m | eta: 37.7m step 03677/05568 (66.04%) | loss: 2.483893 | lrm: 0.55 | dt: 1195.04ms | tok/sec: 877,442 | bf16_mfu: 53.90 | total time: 73.10m | eta: 37.7m step 03678/05568 (66.06%) | loss: 2.485919 | lrm: 0.55 | dt: 1195.80ms | tok/sec: 876,883 | bf16_mfu: 53.86 | total time: 73.12m | eta: 37.7m step 03679/05568 (66.07%) | loss: 2.491898 | lrm: 0.55 | dt: 1190.05ms | tok/sec: 881,122 | bf16_mfu: 54.13 | total time: 73.14m | eta: 37.6m step 03680/05568 (66.09%) | loss: 2.488824 | lrm: 0.55 | dt: 1195.32ms | tok/sec: 877,232 | bf16_mfu: 53.89 | total time: 73.16m | eta: 37.6m step 03681/05568 (66.11%) | loss: 2.484617 | lrm: 0.55 | dt: 1195.20ms | tok/sec: 877,320 | bf16_mfu: 53.89 | total time: 73.18m | eta: 37.6m step 03682/05568 (66.13%) | loss: 2.485306 | lrm: 0.55 | dt: 1192.97ms | tok/sec: 878,965 | bf16_mfu: 53.99 | total time: 73.20m | eta: 37.6m step 03683/05568 (66.15%) | loss: 2.484553 | lrm: 0.54 | dt: 1189.81ms | tok/sec: 881,299 | bf16_mfu: 54.14 | total time: 73.22m | eta: 37.6m step 03684/05568 (66.16%) | loss: 2.494439 | lrm: 0.54 | dt: 1198.34ms | tok/sec: 875,024 | bf16_mfu: 53.75 | total time: 73.24m | eta: 37.5m step 03685/05568 (66.18%) | loss: 2.487671 | lrm: 0.54 | dt: 1195.75ms | tok/sec: 876,922 | bf16_mfu: 53.87 | total time: 73.26m | eta: 37.5m step 03686/05568 (66.20%) | loss: 2.483562 | lrm: 0.54 | dt: 1199.15ms | tok/sec: 874,434 | bf16_mfu: 53.71 | total time: 73.28m | eta: 37.5m step 03687/05568 (66.22%) | loss: 2.484960 | lrm: 0.54 | dt: 1193.23ms | tok/sec: 878,773 | bf16_mfu: 53.98 | total time: 73.30m | eta: 37.5m step 03688/05568 (66.24%) | loss: 2.488236 | lrm: 0.54 | dt: 1192.40ms | tok/sec: 879,379 | bf16_mfu: 54.02 | total time: 73.32m | eta: 37.5m step 03689/05568 (66.25%) | loss: 2.490074 | lrm: 0.54 | dt: 1196.87ms | tok/sec: 876,098 | bf16_mfu: 53.82 | total time: 73.34m | eta: 37.4m step 03690/05568 (66.27%) | loss: 2.493335 | lrm: 0.54 | dt: 1195.95ms | tok/sec: 876,770 | bf16_mfu: 53.86 | total time: 73.36m | eta: 37.4m step 03691/05568 (66.29%) | loss: 2.495675 | lrm: 0.54 | dt: 1192.79ms | tok/sec: 879,093 | bf16_mfu: 54.00 | total time: 73.38m | eta: 37.4m step 03692/05568 (66.31%) | loss: 2.488156 | lrm: 0.54 | dt: 1196.95ms | tok/sec: 876,041 | bf16_mfu: 53.81 | total time: 73.40m | eta: 37.4m step 03693/05568 (66.33%) | loss: 2.493315 | lrm: 0.54 | dt: 1192.29ms | tok/sec: 879,462 | bf16_mfu: 54.02 | total time: 73.42m | eta: 37.4m step 03694/05568 (66.34%) | loss: 2.504976 | lrm: 0.54 | dt: 1194.69ms | tok/sec: 877,693 | bf16_mfu: 53.91 | total time: 73.44m | eta: 37.3m step 03695/05568 (66.36%) | loss: 2.508081 | lrm: 0.54 | dt: 1195.43ms | tok/sec: 877,153 | bf16_mfu: 53.88 | total time: 73.46m | eta: 37.3m step 03696/05568 (66.38%) | loss: 2.515530 | lrm: 0.54 | dt: 1194.46ms | tok/sec: 877,868 | bf16_mfu: 53.93 | total time: 73.48m | eta: 37.3m step 03697/05568 (66.40%) | loss: 2.507177 | lrm: 0.54 | dt: 1196.58ms | tok/sec: 876,309 | bf16_mfu: 53.83 | total time: 73.50m | eta: 37.3m step 03698/05568 (66.42%) | loss: 2.510755 | lrm: 0.54 | dt: 1193.60ms | tok/sec: 878,502 | bf16_mfu: 53.96 | total time: 73.52m | eta: 37.3m step 03699/05568 (66.43%) | loss: 2.506496 | lrm: 0.54 | dt: 1194.20ms | tok/sec: 878,060 | bf16_mfu: 53.94 | total time: 73.54m | eta: 37.2m step 03700/05568 (66.45%) | loss: 2.500281 | lrm: 0.54 | dt: 1198.25ms | tok/sec: 875,091 | bf16_mfu: 53.75 | total time: 73.56m | eta: 37.2m step 03701/05568 (66.47%) | loss: 2.490402 | lrm: 0.54 | dt: 1196.82ms | tok/sec: 876,133 | bf16_mfu: 53.82 | total time: 73.58m | eta: 37.2m step 03702/05568 (66.49%) | loss: 2.489248 | lrm: 0.54 | dt: 1192.14ms | tok/sec: 879,571 | bf16_mfu: 54.03 | total time: 73.60m | eta: 37.2m step 03703/05568 (66.51%) | loss: 2.484330 | lrm: 0.54 | dt: 1193.30ms | tok/sec: 878,721 | bf16_mfu: 53.98 | total time: 73.62m | eta: 37.2m step 03704/05568 (66.52%) | loss: 2.473334 | lrm: 0.54 | dt: 1192.14ms | tok/sec: 879,574 | bf16_mfu: 54.03 | total time: 73.64m | eta: 37.1m step 03705/05568 (66.54%) | loss: 2.467865 | lrm: 0.54 | dt: 1194.74ms | tok/sec: 877,658 | bf16_mfu: 53.91 | total time: 73.66m | eta: 37.1m step 03706/05568 (66.56%) | loss: 2.474650 | lrm: 0.54 | dt: 1198.74ms | tok/sec: 874,734 | bf16_mfu: 53.73 | total time: 73.68m | eta: 37.1m step 03707/05568 (66.58%) | loss: 2.474679 | lrm: 0.54 | dt: 1193.65ms | tok/sec: 878,460 | bf16_mfu: 53.96 | total time: 73.70m | eta: 37.1m step 03708/05568 (66.59%) | loss: 2.465669 | lrm: 0.54 | dt: 1199.07ms | tok/sec: 874,492 | bf16_mfu: 53.72 | total time: 73.72m | eta: 37.1m step 03709/05568 (66.61%) | loss: 2.470555 | lrm: 0.54 | dt: 1196.61ms | tok/sec: 876,289 | bf16_mfu: 53.83 | total time: 73.74m | eta: 37.0m step 03710/05568 (66.63%) | loss: 2.472497 | lrm: 0.54 | dt: 1190.20ms | tok/sec: 881,008 | bf16_mfu: 54.12 | total time: 73.76m | eta: 37.0m step 03711/05568 (66.65%) | loss: 2.464991 | lrm: 0.54 | dt: 1190.28ms | tok/sec: 880,950 | bf16_mfu: 54.11 | total time: 73.78m | eta: 37.0m step 03712/05568 (66.67%) | loss: 2.461813 | lrm: 0.54 | dt: 1192.97ms | tok/sec: 878,959 | bf16_mfu: 53.99 | total time: 73.80m | eta: 37.0m step 03713/05568 (66.68%) | loss: 2.465855 | lrm: 0.54 | dt: 1192.53ms | tok/sec: 879,287 | bf16_mfu: 54.01 | total time: 73.82m | eta: 37.0m step 03714/05568 (66.70%) | loss: 2.472271 | lrm: 0.54 | dt: 1194.72ms | tok/sec: 877,678 | bf16_mfu: 53.91 | total time: 73.84m | eta: 36.9m step 03715/05568 (66.72%) | loss: 2.488874 | lrm: 0.54 | dt: 1198.62ms | tok/sec: 874,819 | bf16_mfu: 53.74 | total time: 73.86m | eta: 36.9m step 03716/05568 (66.74%) | loss: 2.487867 | lrm: 0.54 | dt: 1192.64ms | tok/sec: 879,209 | bf16_mfu: 54.01 | total time: 73.88m | eta: 36.9m step 03717/05568 (66.76%) | loss: 2.481095 | lrm: 0.54 | dt: 1194.35ms | tok/sec: 877,949 | bf16_mfu: 53.93 | total time: 73.90m | eta: 36.9m step 03718/05568 (66.77%) | loss: 2.480251 | lrm: 0.54 | dt: 1195.58ms | tok/sec: 877,042 | bf16_mfu: 53.87 | total time: 73.92m | eta: 36.9m step 03719/05568 (66.79%) | loss: 2.482360 | lrm: 0.54 | dt: 1192.99ms | tok/sec: 878,950 | bf16_mfu: 53.99 | total time: 73.94m | eta: 36.8m step 03720/05568 (66.81%) | loss: 2.481876 | lrm: 0.54 | dt: 1188.73ms | tok/sec: 882,101 | bf16_mfu: 54.19 | total time: 73.96m | eta: 36.8m step 03721/05568 (66.83%) | loss: 2.490693 | lrm: 0.53 | dt: 1194.71ms | tok/sec: 877,682 | bf16_mfu: 53.91 | total time: 73.98m | eta: 36.8m step 03722/05568 (66.85%) | loss: 2.479731 | lrm: 0.53 | dt: 1194.21ms | tok/sec: 878,049 | bf16_mfu: 53.94 | total time: 74.00m | eta: 36.8m step 03723/05568 (66.86%) | loss: 2.480202 | lrm: 0.53 | dt: 1192.14ms | tok/sec: 879,572 | bf16_mfu: 54.03 | total time: 74.02m | eta: 36.8m step 03724/05568 (66.88%) | loss: 2.474912 | lrm: 0.53 | dt: 1198.87ms | tok/sec: 874,640 | bf16_mfu: 53.73 | total time: 74.04m | eta: 36.7m step 03725/05568 (66.90%) | loss: 2.489764 | lrm: 0.53 | dt: 1198.44ms | tok/sec: 874,954 | bf16_mfu: 53.75 | total time: 74.06m | eta: 36.7m step 03726/05568 (66.92%) | loss: 2.496590 | lrm: 0.53 | dt: 1194.95ms | tok/sec: 877,507 | bf16_mfu: 53.90 | total time: 74.08m | eta: 36.7m step 03727/05568 (66.94%) | loss: 2.493059 | lrm: 0.53 | dt: 1193.67ms | tok/sec: 878,445 | bf16_mfu: 53.96 | total time: 74.10m | eta: 36.7m step 03728/05568 (66.95%) | loss: 2.493232 | lrm: 0.53 | dt: 1191.00ms | tok/sec: 880,413 | bf16_mfu: 54.08 | total time: 74.12m | eta: 36.7m step 03729/05568 (66.97%) | loss: 2.492087 | lrm: 0.53 | dt: 1193.99ms | tok/sec: 878,212 | bf16_mfu: 53.95 | total time: 74.14m | eta: 36.6m step 03730/05568 (66.99%) | loss: 2.497249 | lrm: 0.53 | dt: 1194.96ms | tok/sec: 877,501 | bf16_mfu: 53.90 | total time: 74.16m | eta: 36.6m step 03731/05568 (67.01%) | loss: 2.501922 | lrm: 0.53 | dt: 1192.50ms | tok/sec: 879,311 | bf16_mfu: 54.01 | total time: 74.18m | eta: 36.6m step 03732/05568 (67.03%) | loss: 2.490690 | lrm: 0.53 | dt: 1196.23ms | tok/sec: 876,568 | bf16_mfu: 53.85 | total time: 74.20m | eta: 36.6m step 03733/05568 (67.04%) | loss: 2.489372 | lrm: 0.53 | dt: 1194.80ms | tok/sec: 877,616 | bf16_mfu: 53.91 | total time: 74.22m | eta: 36.6m step 03734/05568 (67.06%) | loss: 2.495939 | lrm: 0.53 | dt: 1192.37ms | tok/sec: 879,404 | bf16_mfu: 54.02 | total time: 74.24m | eta: 36.5m step 03735/05568 (67.08%) | loss: 2.502843 | lrm: 0.53 | dt: 1190.67ms | tok/sec: 880,663 | bf16_mfu: 54.10 | total time: 74.26m | eta: 36.5m step 03736/05568 (67.10%) | loss: 2.506327 | lrm: 0.53 | dt: 1196.02ms | tok/sec: 876,722 | bf16_mfu: 53.86 | total time: 74.28m | eta: 36.5m step 03737/05568 (67.12%) | loss: 2.506411 | lrm: 0.53 | dt: 1190.67ms | tok/sec: 880,657 | bf16_mfu: 54.10 | total time: 74.30m | eta: 36.5m step 03738/05568 (67.13%) | loss: 2.498138 | lrm: 0.53 | dt: 1197.61ms | tok/sec: 875,556 | bf16_mfu: 53.78 | total time: 74.32m | eta: 36.5m step 03739/05568 (67.15%) | loss: 2.504239 | lrm: 0.53 | dt: 1198.00ms | tok/sec: 875,270 | bf16_mfu: 53.77 | total time: 74.34m | eta: 36.4m step 03740/05568 (67.17%) | loss: 2.521280 | lrm: 0.53 | dt: 1191.12ms | tok/sec: 880,326 | bf16_mfu: 54.08 | total time: 74.36m | eta: 36.4m step 03741/05568 (67.19%) | loss: 2.521215 | lrm: 0.53 | dt: 1190.14ms | tok/sec: 881,049 | bf16_mfu: 54.12 | total time: 74.38m | eta: 36.4m step 03742/05568 (67.21%) | loss: 2.520413 | lrm: 0.53 | dt: 1191.87ms | tok/sec: 879,774 | bf16_mfu: 54.04 | total time: 74.40m | eta: 36.4m step 03743/05568 (67.22%) | loss: 2.518602 | lrm: 0.53 | dt: 1189.99ms | tok/sec: 881,167 | bf16_mfu: 54.13 | total time: 74.42m | eta: 36.4m step 03744/05568 (67.24%) | loss: 2.508340 | lrm: 0.53 | dt: 1195.71ms | tok/sec: 876,949 | bf16_mfu: 53.87 | total time: 74.44m | eta: 36.3m step 03745/05568 (67.26%) | loss: 2.503905 | lrm: 0.53 | dt: 1193.33ms | tok/sec: 878,696 | bf16_mfu: 53.98 | total time: 74.46m | eta: 36.3m step 03746/05568 (67.28%) | loss: 2.499429 | lrm: 0.53 | dt: 1192.24ms | tok/sec: 879,502 | bf16_mfu: 54.03 | total time: 74.48m | eta: 36.3m step 03747/05568 (67.30%) | loss: 2.501327 | lrm: 0.53 | dt: 1190.21ms | tok/sec: 881,003 | bf16_mfu: 54.12 | total time: 74.50m | eta: 36.3m step 03748/05568 (67.31%) | loss: 2.497631 | lrm: 0.53 | dt: 1196.74ms | tok/sec: 876,196 | bf16_mfu: 53.82 | total time: 74.52m | eta: 36.3m step 03749/05568 (67.33%) | loss: 2.500094 | lrm: 0.53 | dt: 1190.71ms | tok/sec: 880,627 | bf16_mfu: 54.09 | total time: 74.54m | eta: 36.2m step:3750/5568 val_bpb:0.753329 val_time:3.89s step 03750/05568 (67.35%) | loss: 2.493658 | lrm: 0.53 | dt: 1197.19ms | tok/sec: 875,861 | bf16_mfu: 53.80 | total time: 74.56m | eta: 36.2m step 03751/05568 (67.37%) | loss: 2.486822 | lrm: 0.53 | dt: 1196.35ms | tok/sec: 876,481 | bf16_mfu: 53.84 | total time: 74.58m | eta: 36.2m step 03752/05568 (67.39%) | loss: 2.493216 | lrm: 0.53 | dt: 1193.32ms | tok/sec: 878,704 | bf16_mfu: 53.98 | total time: 74.60m | eta: 36.2m step 03753/05568 (67.40%) | loss: 2.491268 | lrm: 0.53 | dt: 1193.20ms | tok/sec: 878,795 | bf16_mfu: 53.98 | total time: 74.62m | eta: 36.2m step 03754/05568 (67.42%) | loss: 2.487448 | lrm: 0.53 | dt: 1198.33ms | tok/sec: 875,027 | bf16_mfu: 53.75 | total time: 74.64m | eta: 36.1m step 03755/05568 (67.44%) | loss: 2.496801 | lrm: 0.53 | dt: 1195.23ms | tok/sec: 877,301 | bf16_mfu: 53.89 | total time: 74.66m | eta: 36.1m step 03756/05568 (67.46%) | loss: 2.507969 | lrm: 0.53 | dt: 1198.81ms | tok/sec: 874,682 | bf16_mfu: 53.73 | total time: 74.68m | eta: 36.1m step 03757/05568 (67.47%) | loss: 2.503844 | lrm: 0.53 | dt: 1192.40ms | tok/sec: 879,386 | bf16_mfu: 54.02 | total time: 74.70m | eta: 36.1m step 03758/05568 (67.49%) | loss: 2.506597 | lrm: 0.53 | dt: 1194.42ms | tok/sec: 877,896 | bf16_mfu: 53.93 | total time: 74.72m | eta: 36.1m step 03759/05568 (67.51%) | loss: 2.501049 | lrm: 0.52 | dt: 1193.85ms | tok/sec: 878,312 | bf16_mfu: 53.95 | total time: 74.74m | eta: 36.0m step 03760/05568 (67.53%) | loss: 2.494785 | lrm: 0.52 | dt: 1196.28ms | tok/sec: 876,530 | bf16_mfu: 53.84 | total time: 74.76m | eta: 36.0m step 03761/05568 (67.55%) | loss: 2.485517 | lrm: 0.52 | dt: 1192.80ms | tok/sec: 879,085 | bf16_mfu: 54.00 | total time: 74.78m | eta: 36.0m step 03762/05568 (67.56%) | loss: 2.477887 | lrm: 0.52 | dt: 1194.05ms | tok/sec: 878,169 | bf16_mfu: 53.94 | total time: 74.80m | eta: 36.0m step 03763/05568 (67.58%) | loss: 2.466930 | lrm: 0.52 | dt: 1196.70ms | tok/sec: 876,226 | bf16_mfu: 53.82 | total time: 74.82m | eta: 36.0m step 03764/05568 (67.60%) | loss: 2.454086 | lrm: 0.52 | dt: 1196.61ms | tok/sec: 876,287 | bf16_mfu: 53.83 | total time: 74.84m | eta: 35.9m step 03765/05568 (67.62%) | loss: 2.449007 | lrm: 0.52 | dt: 1191.83ms | tok/sec: 879,806 | bf16_mfu: 54.04 | total time: 74.86m | eta: 35.9m step 03766/05568 (67.64%) | loss: 2.454087 | lrm: 0.52 | dt: 1193.04ms | tok/sec: 878,914 | bf16_mfu: 53.99 | total time: 74.88m | eta: 35.9m step 03767/05568 (67.65%) | loss: 2.459234 | lrm: 0.52 | dt: 1195.92ms | tok/sec: 876,793 | bf16_mfu: 53.86 | total time: 74.90m | eta: 35.9m step 03768/05568 (67.67%) | loss: 2.467274 | lrm: 0.52 | dt: 1200.48ms | tok/sec: 873,464 | bf16_mfu: 53.65 | total time: 74.92m | eta: 35.9m step 03769/05568 (67.69%) | loss: 2.478291 | lrm: 0.52 | dt: 1193.54ms | tok/sec: 878,541 | bf16_mfu: 53.97 | total time: 74.94m | eta: 35.8m step 03770/05568 (67.71%) | loss: 2.482632 | lrm: 0.52 | dt: 1195.35ms | tok/sec: 877,214 | bf16_mfu: 53.89 | total time: 74.96m | eta: 35.8m step 03771/05568 (67.73%) | loss: 2.490817 | lrm: 0.52 | dt: 1191.14ms | tok/sec: 880,314 | bf16_mfu: 54.08 | total time: 74.98m | eta: 35.8m step 03772/05568 (67.74%) | loss: 2.479424 | lrm: 0.52 | dt: 1197.10ms | tok/sec: 875,926 | bf16_mfu: 53.81 | total time: 75.00m | eta: 35.8m step 03773/05568 (67.76%) | loss: 2.485131 | lrm: 0.52 | dt: 1199.89ms | tok/sec: 873,896 | bf16_mfu: 53.68 | total time: 75.02m | eta: 35.8m step 03774/05568 (67.78%) | loss: 2.483781 | lrm: 0.52 | dt: 1197.14ms | tok/sec: 875,904 | bf16_mfu: 53.80 | total time: 75.04m | eta: 35.7m step 03775/05568 (67.80%) | loss: 2.474024 | lrm: 0.52 | dt: 1196.98ms | tok/sec: 876,020 | bf16_mfu: 53.81 | total time: 75.06m | eta: 35.7m step 03776/05568 (67.82%) | loss: 2.481499 | lrm: 0.52 | dt: 1192.46ms | tok/sec: 879,335 | bf16_mfu: 54.02 | total time: 75.08m | eta: 35.7m step 03777/05568 (67.83%) | loss: 2.494828 | lrm: 0.52 | dt: 1190.86ms | tok/sec: 880,521 | bf16_mfu: 54.09 | total time: 75.09m | eta: 35.7m step 03778/05568 (67.85%) | loss: 2.491182 | lrm: 0.52 | dt: 1195.77ms | tok/sec: 876,903 | bf16_mfu: 53.87 | total time: 75.11m | eta: 35.7m step 03779/05568 (67.87%) | loss: 2.488809 | lrm: 0.52 | dt: 1192.51ms | tok/sec: 879,298 | bf16_mfu: 54.01 | total time: 75.13m | eta: 35.6m step 03780/05568 (67.89%) | loss: 2.487346 | lrm: 0.52 | dt: 1195.38ms | tok/sec: 877,193 | bf16_mfu: 53.88 | total time: 75.15m | eta: 35.6m step 03781/05568 (67.91%) | loss: 2.497476 | lrm: 0.52 | dt: 1196.97ms | tok/sec: 876,027 | bf16_mfu: 53.81 | total time: 75.17m | eta: 35.6m step 03782/05568 (67.92%) | loss: 2.507537 | lrm: 0.52 | dt: 1190.82ms | tok/sec: 880,553 | bf16_mfu: 54.09 | total time: 75.19m | eta: 35.6m step 03783/05568 (67.94%) | loss: 2.502795 | lrm: 0.52 | dt: 1194.69ms | tok/sec: 877,694 | bf16_mfu: 53.91 | total time: 75.21m | eta: 35.6m step 03784/05568 (67.96%) | loss: 2.496778 | lrm: 0.52 | dt: 1195.67ms | tok/sec: 876,974 | bf16_mfu: 53.87 | total time: 75.23m | eta: 35.5m step 03785/05568 (67.98%) | loss: 2.492791 | lrm: 0.52 | dt: 1196.04ms | tok/sec: 876,704 | bf16_mfu: 53.85 | total time: 75.25m | eta: 35.5m step 03786/05568 (68.00%) | loss: 2.504723 | lrm: 0.52 | dt: 1192.23ms | tok/sec: 879,511 | bf16_mfu: 54.03 | total time: 75.27m | eta: 35.5m step 03787/05568 (68.01%) | loss: 2.507569 | lrm: 0.52 | dt: 1194.56ms | tok/sec: 877,789 | bf16_mfu: 53.92 | total time: 75.29m | eta: 35.5m step 03788/05568 (68.03%) | loss: 2.509147 | lrm: 0.52 | dt: 1194.96ms | tok/sec: 877,497 | bf16_mfu: 53.90 | total time: 75.31m | eta: 35.5m step 03789/05568 (68.05%) | loss: 2.521357 | lrm: 0.52 | dt: 1194.80ms | tok/sec: 877,614 | bf16_mfu: 53.91 | total time: 75.33m | eta: 35.4m step 03790/05568 (68.07%) | loss: 2.521115 | lrm: 0.52 | dt: 1197.86ms | tok/sec: 875,371 | bf16_mfu: 53.77 | total time: 75.35m | eta: 35.4m step 03791/05568 (68.09%) | loss: 2.515963 | lrm: 0.52 | dt: 1199.37ms | tok/sec: 874,272 | bf16_mfu: 53.70 | total time: 75.37m | eta: 35.4m step 03792/05568 (68.10%) | loss: 2.500772 | lrm: 0.52 | dt: 1196.57ms | tok/sec: 876,315 | bf16_mfu: 53.83 | total time: 75.39m | eta: 35.4m step 03793/05568 (68.12%) | loss: 2.493880 | lrm: 0.52 | dt: 1197.58ms | tok/sec: 875,576 | bf16_mfu: 53.78 | total time: 75.41m | eta: 35.4m step 03794/05568 (68.14%) | loss: 2.487943 | lrm: 0.52 | dt: 1189.96ms | tok/sec: 881,189 | bf16_mfu: 54.13 | total time: 75.43m | eta: 35.3m step 03795/05568 (68.16%) | loss: 2.494661 | lrm: 0.52 | dt: 1194.22ms | tok/sec: 878,040 | bf16_mfu: 53.94 | total time: 75.45m | eta: 35.3m step 03796/05568 (68.18%) | loss: 2.490314 | lrm: 0.52 | dt: 1194.72ms | tok/sec: 877,676 | bf16_mfu: 53.91 | total time: 75.47m | eta: 35.3m step 03797/05568 (68.19%) | loss: 2.494411 | lrm: 0.51 | dt: 1197.14ms | tok/sec: 875,903 | bf16_mfu: 53.80 | total time: 75.49m | eta: 35.3m step 03798/05568 (68.21%) | loss: 2.487263 | lrm: 0.51 | dt: 1199.43ms | tok/sec: 874,225 | bf16_mfu: 53.70 | total time: 75.51m | eta: 35.3m step 03799/05568 (68.23%) | loss: 2.497190 | lrm: 0.51 | dt: 1194.41ms | tok/sec: 877,904 | bf16_mfu: 53.93 | total time: 75.53m | eta: 35.2m step 03800/05568 (68.25%) | loss: 2.496700 | lrm: 0.51 | dt: 1190.31ms | tok/sec: 880,929 | bf16_mfu: 54.11 | total time: 75.55m | eta: 35.2m step 03801/05568 (68.27%) | loss: 2.498328 | lrm: 0.51 | dt: 1192.64ms | tok/sec: 879,209 | bf16_mfu: 54.01 | total time: 75.57m | eta: 35.2m step 03802/05568 (68.28%) | loss: 2.499236 | lrm: 0.51 | dt: 1194.23ms | tok/sec: 878,038 | bf16_mfu: 53.94 | total time: 75.59m | eta: 35.2m step 03803/05568 (68.30%) | loss: 2.491385 | lrm: 0.51 | dt: 1197.22ms | tok/sec: 875,839 | bf16_mfu: 53.80 | total time: 75.61m | eta: 35.2m step 03804/05568 (68.32%) | loss: 2.482809 | lrm: 0.51 | dt: 1193.35ms | tok/sec: 878,683 | bf16_mfu: 53.98 | total time: 75.63m | eta: 35.1m step 03805/05568 (68.34%) | loss: 2.474238 | lrm: 0.51 | dt: 1191.78ms | tok/sec: 879,841 | bf16_mfu: 54.05 | total time: 75.65m | eta: 35.1m step 03806/05568 (68.35%) | loss: 2.489195 | lrm: 0.51 | dt: 1193.94ms | tok/sec: 878,244 | bf16_mfu: 53.95 | total time: 75.67m | eta: 35.1m step 03807/05568 (68.37%) | loss: 2.482301 | lrm: 0.51 | dt: 1192.21ms | tok/sec: 879,520 | bf16_mfu: 54.03 | total time: 75.69m | eta: 35.1m step 03808/05568 (68.39%) | loss: 2.473146 | lrm: 0.51 | dt: 1197.56ms | tok/sec: 875,592 | bf16_mfu: 53.79 | total time: 75.71m | eta: 35.1m step 03809/05568 (68.41%) | loss: 2.479517 | lrm: 0.51 | dt: 1195.96ms | tok/sec: 876,764 | bf16_mfu: 53.86 | total time: 75.73m | eta: 35.0m step 03810/05568 (68.43%) | loss: 2.472798 | lrm: 0.51 | dt: 1195.04ms | tok/sec: 877,443 | bf16_mfu: 53.90 | total time: 75.75m | eta: 35.0m step 03811/05568 (68.44%) | loss: 2.476208 | lrm: 0.51 | dt: 1195.55ms | tok/sec: 877,067 | bf16_mfu: 53.88 | total time: 75.77m | eta: 35.0m step 03812/05568 (68.46%) | loss: 2.475133 | lrm: 0.51 | dt: 1192.83ms | tok/sec: 879,064 | bf16_mfu: 54.00 | total time: 75.79m | eta: 35.0m step 03813/05568 (68.48%) | loss: 2.490784 | lrm: 0.51 | dt: 1192.96ms | tok/sec: 878,971 | bf16_mfu: 53.99 | total time: 75.81m | eta: 35.0m step 03814/05568 (68.50%) | loss: 2.487031 | lrm: 0.51 | dt: 1198.31ms | tok/sec: 875,046 | bf16_mfu: 53.75 | total time: 75.83m | eta: 34.9m step 03815/05568 (68.52%) | loss: 2.492409 | lrm: 0.51 | dt: 1196.92ms | tok/sec: 876,061 | bf16_mfu: 53.81 | total time: 75.85m | eta: 34.9m step 03816/05568 (68.53%) | loss: 2.492755 | lrm: 0.51 | dt: 1199.64ms | tok/sec: 874,073 | bf16_mfu: 53.69 | total time: 75.87m | eta: 34.9m step 03817/05568 (68.55%) | loss: 2.477777 | lrm: 0.51 | dt: 1190.45ms | tok/sec: 880,822 | bf16_mfu: 54.11 | total time: 75.89m | eta: 34.9m step 03818/05568 (68.57%) | loss: 2.481709 | lrm: 0.51 | dt: 1192.83ms | tok/sec: 879,065 | bf16_mfu: 54.00 | total time: 75.91m | eta: 34.9m step 03819/05568 (68.59%) | loss: 2.484969 | lrm: 0.51 | dt: 1194.61ms | tok/sec: 877,759 | bf16_mfu: 53.92 | total time: 75.93m | eta: 34.8m step 03820/05568 (68.61%) | loss: 2.463311 | lrm: 0.51 | dt: 1190.80ms | tok/sec: 880,564 | bf16_mfu: 54.09 | total time: 75.95m | eta: 34.8m step 03821/05568 (68.62%) | loss: 2.472511 | lrm: 0.51 | dt: 1194.07ms | tok/sec: 878,151 | bf16_mfu: 53.94 | total time: 75.97m | eta: 34.8m step 03822/05568 (68.64%) | loss: 2.463314 | lrm: 0.51 | dt: 1193.20ms | tok/sec: 878,795 | bf16_mfu: 53.98 | total time: 75.99m | eta: 34.8m step 03823/05568 (68.66%) | loss: 2.468402 | lrm: 0.51 | dt: 1193.72ms | tok/sec: 878,407 | bf16_mfu: 53.96 | total time: 76.01m | eta: 34.8m step 03824/05568 (68.68%) | loss: 2.475574 | lrm: 0.51 | dt: 1193.65ms | tok/sec: 878,460 | bf16_mfu: 53.96 | total time: 76.03m | eta: 34.7m step 03825/05568 (68.70%) | loss: 2.475071 | lrm: 0.51 | dt: 1195.63ms | tok/sec: 877,010 | bf16_mfu: 53.87 | total time: 76.05m | eta: 34.7m step 03826/05568 (68.71%) | loss: 2.487758 | lrm: 0.51 | dt: 1198.11ms | tok/sec: 875,192 | bf16_mfu: 53.76 | total time: 76.07m | eta: 34.7m step 03827/05568 (68.73%) | loss: 2.487831 | lrm: 0.51 | dt: 1189.48ms | tok/sec: 881,537 | bf16_mfu: 54.15 | total time: 76.09m | eta: 34.7m step 03828/05568 (68.75%) | loss: 2.487440 | lrm: 0.51 | dt: 1192.59ms | tok/sec: 879,240 | bf16_mfu: 54.01 | total time: 76.11m | eta: 34.7m step 03829/05568 (68.77%) | loss: 2.476107 | lrm: 0.51 | dt: 1193.50ms | tok/sec: 878,570 | bf16_mfu: 53.97 | total time: 76.13m | eta: 34.6m step 03830/05568 (68.79%) | loss: 2.478801 | lrm: 0.51 | dt: 1191.58ms | tok/sec: 879,991 | bf16_mfu: 54.06 | total time: 76.15m | eta: 34.6m step 03831/05568 (68.80%) | loss: 2.467706 | lrm: 0.51 | dt: 1198.17ms | tok/sec: 875,151 | bf16_mfu: 53.76 | total time: 76.17m | eta: 34.6m step 03832/05568 (68.82%) | loss: 2.467912 | lrm: 0.51 | dt: 1201.10ms | tok/sec: 873,010 | bf16_mfu: 53.63 | total time: 76.19m | eta: 34.6m step 03833/05568 (68.84%) | loss: 2.471497 | lrm: 0.51 | dt: 1198.78ms | tok/sec: 874,704 | bf16_mfu: 53.73 | total time: 76.21m | eta: 34.6m step 03834/05568 (68.86%) | loss: 2.470445 | lrm: 0.51 | dt: 1192.48ms | tok/sec: 879,320 | bf16_mfu: 54.01 | total time: 76.23m | eta: 34.5m step 03835/05568 (68.88%) | loss: 2.463146 | lrm: 0.50 | dt: 1193.86ms | tok/sec: 878,305 | bf16_mfu: 53.95 | total time: 76.25m | eta: 34.5m step 03836/05568 (68.89%) | loss: 2.459098 | lrm: 0.50 | dt: 1194.61ms | tok/sec: 877,757 | bf16_mfu: 53.92 | total time: 76.27m | eta: 34.5m step 03837/05568 (68.91%) | loss: 2.461581 | lrm: 0.50 | dt: 1194.52ms | tok/sec: 877,820 | bf16_mfu: 53.92 | total time: 76.29m | eta: 34.5m step 03838/05568 (68.93%) | loss: 2.461771 | lrm: 0.50 | dt: 1192.47ms | tok/sec: 879,330 | bf16_mfu: 54.02 | total time: 76.31m | eta: 34.5m step 03839/05568 (68.95%) | loss: 2.455051 | lrm: 0.50 | dt: 1193.23ms | tok/sec: 878,773 | bf16_mfu: 53.98 | total time: 76.33m | eta: 34.4m step 03840/05568 (68.97%) | loss: 2.454693 | lrm: 0.50 | dt: 1193.78ms | tok/sec: 878,364 | bf16_mfu: 53.96 | total time: 76.35m | eta: 34.4m step 03841/05568 (68.98%) | loss: 2.450264 | lrm: 0.50 | dt: 1195.33ms | tok/sec: 877,227 | bf16_mfu: 53.89 | total time: 76.37m | eta: 34.4m step 03842/05568 (69.00%) | loss: 2.452690 | lrm: 0.50 | dt: 1192.50ms | tok/sec: 879,308 | bf16_mfu: 54.01 | total time: 76.39m | eta: 34.4m step 03843/05568 (69.02%) | loss: 2.461699 | lrm: 0.50 | dt: 1193.72ms | tok/sec: 878,408 | bf16_mfu: 53.96 | total time: 76.41m | eta: 34.4m step 03844/05568 (69.04%) | loss: 2.461624 | lrm: 0.50 | dt: 1194.48ms | tok/sec: 877,854 | bf16_mfu: 53.92 | total time: 76.43m | eta: 34.3m step 03845/05568 (69.06%) | loss: 2.468232 | lrm: 0.50 | dt: 1192.69ms | tok/sec: 879,167 | bf16_mfu: 54.01 | total time: 76.45m | eta: 34.3m step 03846/05568 (69.07%) | loss: 2.463529 | lrm: 0.50 | dt: 1195.52ms | tok/sec: 877,090 | bf16_mfu: 53.88 | total time: 76.47m | eta: 34.3m step 03847/05568 (69.09%) | loss: 2.462206 | lrm: 0.50 | dt: 1203.09ms | tok/sec: 871,571 | bf16_mfu: 53.54 | total time: 76.49m | eta: 34.3m step 03848/05568 (69.11%) | loss: 2.464962 | lrm: 0.50 | dt: 1193.56ms | tok/sec: 878,525 | bf16_mfu: 53.97 | total time: 76.51m | eta: 34.3m step 03849/05568 (69.13%) | loss: 2.473457 | lrm: 0.50 | dt: 1191.94ms | tok/sec: 879,719 | bf16_mfu: 54.04 | total time: 76.53m | eta: 34.2m step 03850/05568 (69.15%) | loss: 2.466283 | lrm: 0.50 | dt: 1196.42ms | tok/sec: 876,429 | bf16_mfu: 53.84 | total time: 76.55m | eta: 34.2m step 03851/05568 (69.16%) | loss: 2.453564 | lrm: 0.50 | dt: 1193.45ms | tok/sec: 878,611 | bf16_mfu: 53.97 | total time: 76.57m | eta: 34.2m step 03852/05568 (69.18%) | loss: 2.443426 | lrm: 0.50 | dt: 1194.37ms | tok/sec: 877,928 | bf16_mfu: 53.93 | total time: 76.59m | eta: 34.2m step 03853/05568 (69.20%) | loss: 2.440300 | lrm: 0.50 | dt: 1196.01ms | tok/sec: 876,728 | bf16_mfu: 53.86 | total time: 76.61m | eta: 34.2m step 03854/05568 (69.22%) | loss: 2.438343 | lrm: 0.50 | dt: 1198.92ms | tok/sec: 874,601 | bf16_mfu: 53.72 | total time: 76.63m | eta: 34.1m step 03855/05568 (69.23%) | loss: 2.446459 | lrm: 0.50 | dt: 1193.87ms | tok/sec: 878,298 | bf16_mfu: 53.95 | total time: 76.65m | eta: 34.1m step 03856/05568 (69.25%) | loss: 2.443488 | lrm: 0.50 | dt: 1191.80ms | tok/sec: 879,827 | bf16_mfu: 54.05 | total time: 76.67m | eta: 34.1m step 03857/05568 (69.27%) | loss: 2.445666 | lrm: 0.50 | dt: 1195.94ms | tok/sec: 876,782 | bf16_mfu: 53.86 | total time: 76.69m | eta: 34.1m step 03858/05568 (69.29%) | loss: 2.441576 | lrm: 0.50 | dt: 1197.93ms | tok/sec: 875,319 | bf16_mfu: 53.77 | total time: 76.71m | eta: 34.1m step 03859/05568 (69.31%) | loss: 2.444176 | lrm: 0.50 | dt: 1197.78ms | tok/sec: 875,434 | bf16_mfu: 53.78 | total time: 76.73m | eta: 34.0m step 03860/05568 (69.32%) | loss: 2.445817 | lrm: 0.50 | dt: 1190.19ms | tok/sec: 881,017 | bf16_mfu: 54.12 | total time: 76.75m | eta: 34.0m step 03861/05568 (69.34%) | loss: 2.445605 | lrm: 0.50 | dt: 1188.94ms | tok/sec: 881,938 | bf16_mfu: 54.18 | total time: 76.77m | eta: 34.0m step 03862/05568 (69.36%) | loss: 2.458223 | lrm: 0.50 | dt: 1192.87ms | tok/sec: 879,037 | bf16_mfu: 54.00 | total time: 76.79m | eta: 34.0m step 03863/05568 (69.38%) | loss: 2.458620 | lrm: 0.50 | dt: 1190.40ms | tok/sec: 880,861 | bf16_mfu: 54.11 | total time: 76.81m | eta: 34.0m step 03864/05568 (69.40%) | loss: 2.456365 | lrm: 0.50 | dt: 1190.58ms | tok/sec: 880,730 | bf16_mfu: 54.10 | total time: 76.83m | eta: 33.9m step 03865/05568 (69.41%) | loss: 2.449608 | lrm: 0.50 | dt: 1194.75ms | tok/sec: 877,655 | bf16_mfu: 53.91 | total time: 76.85m | eta: 33.9m step 03866/05568 (69.43%) | loss: 2.459938 | lrm: 0.50 | dt: 1194.47ms | tok/sec: 877,857 | bf16_mfu: 53.92 | total time: 76.87m | eta: 33.9m step 03867/05568 (69.45%) | loss: 2.463082 | lrm: 0.50 | dt: 1193.25ms | tok/sec: 878,757 | bf16_mfu: 53.98 | total time: 76.89m | eta: 33.9m step 03868/05568 (69.47%) | loss: 2.465356 | lrm: 0.50 | dt: 1196.21ms | tok/sec: 876,583 | bf16_mfu: 53.85 | total time: 76.91m | eta: 33.9m step 03869/05568 (69.49%) | loss: 2.475441 | lrm: 0.50 | dt: 1194.05ms | tok/sec: 878,165 | bf16_mfu: 53.94 | total time: 76.93m | eta: 33.8m step 03870/05568 (69.50%) | loss: 2.469804 | lrm: 0.50 | dt: 1195.05ms | tok/sec: 877,431 | bf16_mfu: 53.90 | total time: 76.95m | eta: 33.8m step 03871/05568 (69.52%) | loss: 2.469028 | lrm: 0.50 | dt: 1196.70ms | tok/sec: 876,224 | bf16_mfu: 53.82 | total time: 76.97m | eta: 33.8m step 03872/05568 (69.54%) | loss: 2.469403 | lrm: 0.50 | dt: 1195.05ms | tok/sec: 877,432 | bf16_mfu: 53.90 | total time: 76.99m | eta: 33.8m step 03873/05568 (69.56%) | loss: 2.469844 | lrm: 0.49 | dt: 1194.73ms | tok/sec: 877,669 | bf16_mfu: 53.91 | total time: 77.01m | eta: 33.8m step 03874/05568 (69.58%) | loss: 2.456516 | lrm: 0.49 | dt: 1193.89ms | tok/sec: 878,284 | bf16_mfu: 53.95 | total time: 77.03m | eta: 33.7m step 03875/05568 (69.59%) | loss: 2.455083 | lrm: 0.49 | dt: 1199.71ms | tok/sec: 874,024 | bf16_mfu: 53.69 | total time: 77.05m | eta: 33.7m step 03876/05568 (69.61%) | loss: 2.464663 | lrm: 0.49 | dt: 1191.50ms | tok/sec: 880,047 | bf16_mfu: 54.06 | total time: 77.07m | eta: 33.7m step 03877/05568 (69.63%) | loss: 2.468858 | lrm: 0.49 | dt: 1190.48ms | tok/sec: 880,804 | bf16_mfu: 54.11 | total time: 77.09m | eta: 33.7m step 03878/05568 (69.65%) | loss: 2.484336 | lrm: 0.49 | dt: 1192.86ms | tok/sec: 879,043 | bf16_mfu: 54.00 | total time: 77.11m | eta: 33.7m step 03879/05568 (69.67%) | loss: 2.475172 | lrm: 0.49 | dt: 1196.94ms | tok/sec: 876,048 | bf16_mfu: 53.81 | total time: 77.13m | eta: 33.6m step 03880/05568 (69.68%) | loss: 2.480404 | lrm: 0.49 | dt: 1195.16ms | tok/sec: 877,352 | bf16_mfu: 53.89 | total time: 77.15m | eta: 33.6m step 03881/05568 (69.70%) | loss: 2.486626 | lrm: 0.49 | dt: 1193.50ms | tok/sec: 878,570 | bf16_mfu: 53.97 | total time: 77.17m | eta: 33.6m step 03882/05568 (69.72%) | loss: 2.481132 | lrm: 0.49 | dt: 1197.96ms | tok/sec: 875,303 | bf16_mfu: 53.77 | total time: 77.19m | eta: 33.6m step 03883/05568 (69.74%) | loss: 2.473208 | lrm: 0.49 | dt: 1192.15ms | tok/sec: 879,568 | bf16_mfu: 54.03 | total time: 77.21m | eta: 33.6m step 03884/05568 (69.76%) | loss: 2.471097 | lrm: 0.49 | dt: 1201.41ms | tok/sec: 872,787 | bf16_mfu: 53.61 | total time: 77.23m | eta: 33.5m step 03885/05568 (69.77%) | loss: 2.454577 | lrm: 0.49 | dt: 1193.72ms | tok/sec: 878,411 | bf16_mfu: 53.96 | total time: 77.25m | eta: 33.5m step 03886/05568 (69.79%) | loss: 2.455149 | lrm: 0.49 | dt: 1192.56ms | tok/sec: 879,266 | bf16_mfu: 54.01 | total time: 77.27m | eta: 33.5m step 03887/05568 (69.81%) | loss: 2.452656 | lrm: 0.49 | dt: 1192.30ms | tok/sec: 879,456 | bf16_mfu: 54.02 | total time: 77.29m | eta: 33.5m step 03888/05568 (69.83%) | loss: 2.443565 | lrm: 0.49 | dt: 1195.59ms | tok/sec: 877,038 | bf16_mfu: 53.87 | total time: 77.30m | eta: 33.5m step 03889/05568 (69.85%) | loss: 2.457744 | lrm: 0.49 | dt: 1194.06ms | tok/sec: 878,161 | bf16_mfu: 53.94 | total time: 77.32m | eta: 33.4m step 03890/05568 (69.86%) | loss: 2.468893 | lrm: 0.49 | dt: 1201.06ms | tok/sec: 873,041 | bf16_mfu: 53.63 | total time: 77.34m | eta: 33.4m step 03891/05568 (69.88%) | loss: 2.475442 | lrm: 0.49 | dt: 1195.61ms | tok/sec: 877,019 | bf16_mfu: 53.87 | total time: 77.36m | eta: 33.4m step 03892/05568 (69.90%) | loss: 2.482960 | lrm: 0.49 | dt: 1194.37ms | tok/sec: 877,931 | bf16_mfu: 53.93 | total time: 77.38m | eta: 33.4m step 03893/05568 (69.92%) | loss: 2.494186 | lrm: 0.49 | dt: 1194.48ms | tok/sec: 877,849 | bf16_mfu: 53.92 | total time: 77.40m | eta: 33.4m step 03894/05568 (69.94%) | loss: 2.492548 | lrm: 0.49 | dt: 1191.53ms | tok/sec: 880,023 | bf16_mfu: 54.06 | total time: 77.42m | eta: 33.3m step 03895/05568 (69.95%) | loss: 2.489387 | lrm: 0.49 | dt: 1192.47ms | tok/sec: 879,331 | bf16_mfu: 54.02 | total time: 77.44m | eta: 33.3m step 03896/05568 (69.97%) | loss: 2.497870 | lrm: 0.49 | dt: 1196.98ms | tok/sec: 876,015 | bf16_mfu: 53.81 | total time: 77.46m | eta: 33.3m step 03897/05568 (69.99%) | loss: 2.498319 | lrm: 0.49 | dt: 1196.43ms | tok/sec: 876,424 | bf16_mfu: 53.84 | total time: 77.48m | eta: 33.3m step 03898/05568 (70.01%) | loss: 2.486192 | lrm: 0.49 | dt: 1193.87ms | tok/sec: 878,298 | bf16_mfu: 53.95 | total time: 77.50m | eta: 33.3m step 03899/05568 (70.03%) | loss: 2.478954 | lrm: 0.49 | dt: 1195.79ms | tok/sec: 876,893 | bf16_mfu: 53.87 | total time: 77.52m | eta: 33.3m step 03900/05568 (70.04%) | loss: 2.481304 | lrm: 0.49 | dt: 1196.81ms | tok/sec: 876,146 | bf16_mfu: 53.82 | total time: 77.54m | eta: 33.2m step 03901/05568 (70.06%) | loss: 2.480127 | lrm: 0.49 | dt: 1197.04ms | tok/sec: 875,971 | bf16_mfu: 53.81 | total time: 77.56m | eta: 33.2m step 03902/05568 (70.08%) | loss: 2.471259 | lrm: 0.49 | dt: 1193.41ms | tok/sec: 878,639 | bf16_mfu: 53.97 | total time: 77.58m | eta: 33.2m step 03903/05568 (70.10%) | loss: 2.470438 | lrm: 0.49 | dt: 1197.38ms | tok/sec: 875,726 | bf16_mfu: 53.79 | total time: 77.60m | eta: 33.2m step 03904/05568 (70.11%) | loss: 2.468078 | lrm: 0.49 | dt: 1194.07ms | tok/sec: 878,151 | bf16_mfu: 53.94 | total time: 77.62m | eta: 33.2m step 03905/05568 (70.13%) | loss: 2.465079 | lrm: 0.49 | dt: 1196.74ms | tok/sec: 876,192 | bf16_mfu: 53.82 | total time: 77.64m | eta: 33.1m step 03906/05568 (70.15%) | loss: 2.467250 | lrm: 0.49 | dt: 1194.40ms | tok/sec: 877,910 | bf16_mfu: 53.93 | total time: 77.66m | eta: 33.1m step 03907/05568 (70.17%) | loss: 2.454725 | lrm: 0.49 | dt: 1190.87ms | tok/sec: 880,510 | bf16_mfu: 54.09 | total time: 77.68m | eta: 33.1m step 03908/05568 (70.19%) | loss: 2.466532 | lrm: 0.49 | dt: 1191.78ms | tok/sec: 879,837 | bf16_mfu: 54.05 | total time: 77.70m | eta: 33.1m step 03909/05568 (70.20%) | loss: 2.466696 | lrm: 0.49 | dt: 1195.40ms | tok/sec: 877,178 | bf16_mfu: 53.88 | total time: 77.72m | eta: 33.1m step 03910/05568 (70.22%) | loss: 2.468208 | lrm: 0.49 | dt: 1194.28ms | tok/sec: 877,995 | bf16_mfu: 53.93 | total time: 77.74m | eta: 33.0m step 03911/05568 (70.24%) | loss: 2.456530 | lrm: 0.48 | dt: 1193.70ms | tok/sec: 878,426 | bf16_mfu: 53.96 | total time: 77.76m | eta: 33.0m step 03912/05568 (70.26%) | loss: 2.452674 | lrm: 0.48 | dt: 1195.64ms | tok/sec: 877,001 | bf16_mfu: 53.87 | total time: 77.78m | eta: 33.0m step 03913/05568 (70.28%) | loss: 2.453669 | lrm: 0.48 | dt: 1194.63ms | tok/sec: 877,738 | bf16_mfu: 53.92 | total time: 77.80m | eta: 33.0m step 03914/05568 (70.29%) | loss: 2.445452 | lrm: 0.48 | dt: 1198.21ms | tok/sec: 875,115 | bf16_mfu: 53.76 | total time: 77.82m | eta: 33.0m step 03915/05568 (70.31%) | loss: 2.446131 | lrm: 0.48 | dt: 1194.80ms | tok/sec: 877,613 | bf16_mfu: 53.91 | total time: 77.84m | eta: 32.9m step 03916/05568 (70.33%) | loss: 2.445120 | lrm: 0.48 | dt: 1195.52ms | tok/sec: 877,091 | bf16_mfu: 53.88 | total time: 77.86m | eta: 32.9m step 03917/05568 (70.35%) | loss: 2.447432 | lrm: 0.48 | dt: 1194.95ms | tok/sec: 877,506 | bf16_mfu: 53.90 | total time: 77.88m | eta: 32.9m step 03918/05568 (70.37%) | loss: 2.448287 | lrm: 0.48 | dt: 1191.69ms | tok/sec: 879,904 | bf16_mfu: 54.05 | total time: 77.90m | eta: 32.9m step 03919/05568 (70.38%) | loss: 2.445072 | lrm: 0.48 | dt: 1192.94ms | tok/sec: 878,985 | bf16_mfu: 53.99 | total time: 77.92m | eta: 32.9m step 03920/05568 (70.40%) | loss: 2.437551 | lrm: 0.48 | dt: 1194.65ms | tok/sec: 877,725 | bf16_mfu: 53.92 | total time: 77.94m | eta: 32.8m step 03921/05568 (70.42%) | loss: 2.427638 | lrm: 0.48 | dt: 1198.50ms | tok/sec: 874,909 | bf16_mfu: 53.74 | total time: 77.96m | eta: 32.8m step 03922/05568 (70.44%) | loss: 2.439063 | lrm: 0.48 | dt: 1188.89ms | tok/sec: 881,981 | bf16_mfu: 54.18 | total time: 77.98m | eta: 32.8m step 03923/05568 (70.46%) | loss: 2.435172 | lrm: 0.48 | dt: 1196.71ms | tok/sec: 876,217 | bf16_mfu: 53.82 | total time: 78.00m | eta: 32.8m step 03924/05568 (70.47%) | loss: 2.438758 | lrm: 0.48 | dt: 1193.86ms | tok/sec: 878,310 | bf16_mfu: 53.95 | total time: 78.02m | eta: 32.8m step 03925/05568 (70.49%) | loss: 2.450305 | lrm: 0.48 | dt: 1194.84ms | tok/sec: 877,589 | bf16_mfu: 53.91 | total time: 78.04m | eta: 32.7m step 03926/05568 (70.51%) | loss: 2.445315 | lrm: 0.48 | dt: 1194.96ms | tok/sec: 877,501 | bf16_mfu: 53.90 | total time: 78.06m | eta: 32.7m step 03927/05568 (70.53%) | loss: 2.434039 | lrm: 0.48 | dt: 1195.84ms | tok/sec: 876,853 | bf16_mfu: 53.86 | total time: 78.08m | eta: 32.7m step 03928/05568 (70.55%) | loss: 2.431940 | lrm: 0.48 | dt: 1192.62ms | tok/sec: 879,223 | bf16_mfu: 54.01 | total time: 78.10m | eta: 32.7m step 03929/05568 (70.56%) | loss: 2.427519 | lrm: 0.48 | dt: 1196.78ms | tok/sec: 876,161 | bf16_mfu: 53.82 | total time: 78.12m | eta: 32.7m step 03930/05568 (70.58%) | loss: 2.443779 | lrm: 0.48 | dt: 1201.53ms | tok/sec: 872,697 | bf16_mfu: 53.61 | total time: 78.14m | eta: 32.6m step 03931/05568 (70.60%) | loss: 2.447469 | lrm: 0.48 | dt: 1192.49ms | tok/sec: 879,315 | bf16_mfu: 54.01 | total time: 78.16m | eta: 32.6m step 03932/05568 (70.62%) | loss: 2.450882 | lrm: 0.48 | dt: 1200.43ms | tok/sec: 873,502 | bf16_mfu: 53.66 | total time: 78.18m | eta: 32.6m step 03933/05568 (70.64%) | loss: 2.455139 | lrm: 0.48 | dt: 1192.23ms | tok/sec: 879,507 | bf16_mfu: 54.03 | total time: 78.20m | eta: 32.6m step 03934/05568 (70.65%) | loss: 2.459719 | lrm: 0.48 | dt: 1193.32ms | tok/sec: 878,702 | bf16_mfu: 53.98 | total time: 78.22m | eta: 32.6m step 03935/05568 (70.67%) | loss: 2.471120 | lrm: 0.48 | dt: 1193.34ms | tok/sec: 878,689 | bf16_mfu: 53.98 | total time: 78.24m | eta: 32.5m step 03936/05568 (70.69%) | loss: 2.485002 | lrm: 0.48 | dt: 1196.77ms | tok/sec: 876,172 | bf16_mfu: 53.82 | total time: 78.26m | eta: 32.5m step 03937/05568 (70.71%) | loss: 2.490955 | lrm: 0.48 | dt: 1197.24ms | tok/sec: 875,825 | bf16_mfu: 53.80 | total time: 78.28m | eta: 32.5m step 03938/05568 (70.73%) | loss: 2.487423 | lrm: 0.48 | dt: 1194.98ms | tok/sec: 877,484 | bf16_mfu: 53.90 | total time: 78.30m | eta: 32.5m step 03939/05568 (70.74%) | loss: 2.482342 | lrm: 0.48 | dt: 1195.09ms | tok/sec: 877,401 | bf16_mfu: 53.90 | total time: 78.32m | eta: 32.5m step 03940/05568 (70.76%) | loss: 2.477309 | lrm: 0.48 | dt: 1194.70ms | tok/sec: 877,690 | bf16_mfu: 53.91 | total time: 78.34m | eta: 32.4m step 03941/05568 (70.78%) | loss: 2.475745 | lrm: 0.48 | dt: 1196.66ms | tok/sec: 876,251 | bf16_mfu: 53.83 | total time: 78.36m | eta: 32.4m step 03942/05568 (70.80%) | loss: 2.465406 | lrm: 0.48 | dt: 1193.60ms | tok/sec: 878,499 | bf16_mfu: 53.96 | total time: 78.38m | eta: 32.4m step 03943/05568 (70.82%) | loss: 2.469383 | lrm: 0.48 | dt: 1191.12ms | tok/sec: 880,331 | bf16_mfu: 54.08 | total time: 78.40m | eta: 32.4m step 03944/05568 (70.83%) | loss: 2.461977 | lrm: 0.48 | dt: 1192.31ms | tok/sec: 879,446 | bf16_mfu: 54.02 | total time: 78.42m | eta: 32.4m step 03945/05568 (70.85%) | loss: 2.467602 | lrm: 0.48 | dt: 1194.31ms | tok/sec: 877,977 | bf16_mfu: 53.93 | total time: 78.44m | eta: 32.3m step 03946/05568 (70.87%) | loss: 2.457966 | lrm: 0.48 | dt: 1195.98ms | tok/sec: 876,752 | bf16_mfu: 53.86 | total time: 78.46m | eta: 32.3m step 03947/05568 (70.89%) | loss: 2.455899 | lrm: 0.48 | dt: 1194.31ms | tok/sec: 877,979 | bf16_mfu: 53.93 | total time: 78.48m | eta: 32.3m step 03948/05568 (70.91%) | loss: 2.466688 | lrm: 0.48 | dt: 1197.59ms | tok/sec: 875,571 | bf16_mfu: 53.78 | total time: 78.50m | eta: 32.3m step 03949/05568 (70.92%) | loss: 2.469487 | lrm: 0.47 | dt: 1193.42ms | tok/sec: 878,632 | bf16_mfu: 53.97 | total time: 78.52m | eta: 32.3m step 03950/05568 (70.94%) | loss: 2.462290 | lrm: 0.47 | dt: 1194.45ms | tok/sec: 877,871 | bf16_mfu: 53.93 | total time: 78.54m | eta: 32.2m step 03951/05568 (70.96%) | loss: 2.471789 | lrm: 0.47 | dt: 1200.60ms | tok/sec: 873,375 | bf16_mfu: 53.65 | total time: 78.56m | eta: 32.2m step 03952/05568 (70.98%) | loss: 2.474576 | lrm: 0.47 | dt: 1194.79ms | tok/sec: 877,625 | bf16_mfu: 53.91 | total time: 78.58m | eta: 32.2m step 03953/05568 (70.99%) | loss: 2.469796 | lrm: 0.47 | dt: 1201.98ms | tok/sec: 872,370 | bf16_mfu: 53.59 | total time: 78.60m | eta: 32.2m step 03954/05568 (71.01%) | loss: 2.459852 | lrm: 0.47 | dt: 1193.09ms | tok/sec: 878,872 | bf16_mfu: 53.99 | total time: 78.62m | eta: 32.2m step 03955/05568 (71.03%) | loss: 2.450518 | lrm: 0.47 | dt: 1197.84ms | tok/sec: 875,392 | bf16_mfu: 53.77 | total time: 78.64m | eta: 32.1m step 03956/05568 (71.05%) | loss: 2.452878 | lrm: 0.47 | dt: 1197.26ms | tok/sec: 875,812 | bf16_mfu: 53.80 | total time: 78.66m | eta: 32.1m step 03957/05568 (71.07%) | loss: 2.459969 | lrm: 0.47 | dt: 1192.08ms | tok/sec: 879,618 | bf16_mfu: 54.03 | total time: 78.68m | eta: 32.1m step 03958/05568 (71.08%) | loss: 2.459750 | lrm: 0.47 | dt: 1193.87ms | tok/sec: 878,302 | bf16_mfu: 53.95 | total time: 78.70m | eta: 32.1m step 03959/05568 (71.10%) | loss: 2.455073 | lrm: 0.47 | dt: 1189.74ms | tok/sec: 881,349 | bf16_mfu: 54.14 | total time: 78.72m | eta: 32.1m step 03960/05568 (71.12%) | loss: 2.460132 | lrm: 0.47 | dt: 1195.03ms | tok/sec: 877,445 | bf16_mfu: 53.90 | total time: 78.74m | eta: 32.0m step 03961/05568 (71.14%) | loss: 2.463157 | lrm: 0.47 | dt: 1195.13ms | tok/sec: 877,370 | bf16_mfu: 53.89 | total time: 78.76m | eta: 32.0m step 03962/05568 (71.16%) | loss: 2.454931 | lrm: 0.47 | dt: 1191.63ms | tok/sec: 879,948 | bf16_mfu: 54.05 | total time: 78.78m | eta: 32.0m step 03963/05568 (71.17%) | loss: 2.463625 | lrm: 0.47 | dt: 1193.13ms | tok/sec: 878,841 | bf16_mfu: 53.99 | total time: 78.80m | eta: 32.0m step 03964/05568 (71.19%) | loss: 2.466118 | lrm: 0.47 | dt: 1195.24ms | tok/sec: 877,290 | bf16_mfu: 53.89 | total time: 78.82m | eta: 32.0m step 03965/05568 (71.21%) | loss: 2.453977 | lrm: 0.47 | dt: 1195.56ms | tok/sec: 877,058 | bf16_mfu: 53.88 | total time: 78.84m | eta: 31.9m step 03966/05568 (71.23%) | loss: 2.459188 | lrm: 0.47 | dt: 1192.47ms | tok/sec: 879,329 | bf16_mfu: 54.02 | total time: 78.86m | eta: 31.9m step 03967/05568 (71.25%) | loss: 2.449386 | lrm: 0.47 | dt: 1193.43ms | tok/sec: 878,624 | bf16_mfu: 53.97 | total time: 78.88m | eta: 31.9m step 03968/05568 (71.26%) | loss: 2.451689 | lrm: 0.47 | dt: 1196.92ms | tok/sec: 876,063 | bf16_mfu: 53.81 | total time: 78.90m | eta: 31.9m step 03969/05568 (71.28%) | loss: 2.456278 | lrm: 0.47 | dt: 1195.42ms | tok/sec: 877,161 | bf16_mfu: 53.88 | total time: 78.92m | eta: 31.9m step 03970/05568 (71.30%) | loss: 2.459422 | lrm: 0.47 | dt: 1199.73ms | tok/sec: 874,007 | bf16_mfu: 53.69 | total time: 78.94m | eta: 31.8m step 03971/05568 (71.32%) | loss: 2.451857 | lrm: 0.47 | dt: 1193.60ms | tok/sec: 878,499 | bf16_mfu: 53.96 | total time: 78.96m | eta: 31.8m step 03972/05568 (71.34%) | loss: 2.447148 | lrm: 0.47 | dt: 1193.40ms | tok/sec: 878,646 | bf16_mfu: 53.97 | total time: 78.98m | eta: 31.8m step 03973/05568 (71.35%) | loss: 2.439857 | lrm: 0.47 | dt: 1194.28ms | tok/sec: 877,997 | bf16_mfu: 53.93 | total time: 79.00m | eta: 31.8m step 03974/05568 (71.37%) | loss: 2.441826 | lrm: 0.47 | dt: 1194.00ms | tok/sec: 878,204 | bf16_mfu: 53.95 | total time: 79.02m | eta: 31.8m step 03975/05568 (71.39%) | loss: 2.431612 | lrm: 0.47 | dt: 1195.54ms | tok/sec: 877,072 | bf16_mfu: 53.88 | total time: 79.04m | eta: 31.7m step 03976/05568 (71.41%) | loss: 2.445680 | lrm: 0.47 | dt: 1192.76ms | tok/sec: 879,114 | bf16_mfu: 54.00 | total time: 79.06m | eta: 31.7m step 03977/05568 (71.43%) | loss: 2.445076 | lrm: 0.47 | dt: 1197.31ms | tok/sec: 875,778 | bf16_mfu: 53.80 | total time: 79.08m | eta: 31.7m step 03978/05568 (71.44%) | loss: 2.438104 | lrm: 0.47 | dt: 1197.96ms | tok/sec: 875,299 | bf16_mfu: 53.77 | total time: 79.10m | eta: 31.7m step 03979/05568 (71.46%) | loss: 2.437275 | lrm: 0.47 | dt: 1191.37ms | tok/sec: 880,139 | bf16_mfu: 54.06 | total time: 79.12m | eta: 31.7m step 03980/05568 (71.48%) | loss: 2.437168 | lrm: 0.47 | dt: 1190.83ms | tok/sec: 880,539 | bf16_mfu: 54.09 | total time: 79.14m | eta: 31.6m step 03981/05568 (71.50%) | loss: 2.438714 | lrm: 0.47 | dt: 1196.63ms | tok/sec: 876,271 | bf16_mfu: 53.83 | total time: 79.16m | eta: 31.6m step 03982/05568 (71.52%) | loss: 2.438560 | lrm: 0.47 | dt: 1202.94ms | tok/sec: 871,675 | bf16_mfu: 53.55 | total time: 79.18m | eta: 31.6m step 03983/05568 (71.53%) | loss: 2.445279 | lrm: 0.47 | dt: 1198.16ms | tok/sec: 875,158 | bf16_mfu: 53.76 | total time: 79.20m | eta: 31.6m step 03984/05568 (71.55%) | loss: 2.455523 | lrm: 0.47 | dt: 1193.55ms | tok/sec: 878,536 | bf16_mfu: 53.97 | total time: 79.22m | eta: 31.6m step 03985/05568 (71.57%) | loss: 2.457468 | lrm: 0.47 | dt: 1192.10ms | tok/sec: 879,606 | bf16_mfu: 54.03 | total time: 79.24m | eta: 31.5m step 03986/05568 (71.59%) | loss: 2.448053 | lrm: 0.47 | dt: 1197.10ms | tok/sec: 875,932 | bf16_mfu: 53.81 | total time: 79.26m | eta: 31.5m step 03987/05568 (71.61%) | loss: 2.443062 | lrm: 0.47 | dt: 1191.95ms | tok/sec: 879,713 | bf16_mfu: 54.04 | total time: 79.28m | eta: 31.5m step 03988/05568 (71.62%) | loss: 2.446293 | lrm: 0.46 | dt: 1194.75ms | tok/sec: 877,655 | bf16_mfu: 53.91 | total time: 79.30m | eta: 31.5m step 03989/05568 (71.64%) | loss: 2.457183 | lrm: 0.46 | dt: 1193.96ms | tok/sec: 878,230 | bf16_mfu: 53.95 | total time: 79.32m | eta: 31.5m step 03990/05568 (71.66%) | loss: 2.458970 | lrm: 0.46 | dt: 1195.75ms | tok/sec: 876,918 | bf16_mfu: 53.87 | total time: 79.34m | eta: 31.4m step 03991/05568 (71.68%) | loss: 2.459566 | lrm: 0.46 | dt: 1192.71ms | tok/sec: 879,157 | bf16_mfu: 54.00 | total time: 79.36m | eta: 31.4m step 03992/05568 (71.70%) | loss: 2.463967 | lrm: 0.46 | dt: 1194.21ms | tok/sec: 878,052 | bf16_mfu: 53.94 | total time: 79.38m | eta: 31.4m step 03993/05568 (71.71%) | loss: 2.469377 | lrm: 0.46 | dt: 1198.75ms | tok/sec: 874,725 | bf16_mfu: 53.73 | total time: 79.40m | eta: 31.4m step 03994/05568 (71.73%) | loss: 2.485150 | lrm: 0.46 | dt: 1193.05ms | tok/sec: 878,903 | bf16_mfu: 53.99 | total time: 79.42m | eta: 31.4m step 03995/05568 (71.75%) | loss: 2.474117 | lrm: 0.46 | dt: 1191.94ms | tok/sec: 879,725 | bf16_mfu: 54.04 | total time: 79.44m | eta: 31.3m step 03996/05568 (71.77%) | loss: 2.466695 | lrm: 0.46 | dt: 1199.05ms | tok/sec: 874,509 | bf16_mfu: 53.72 | total time: 79.46m | eta: 31.3m step 03997/05568 (71.79%) | loss: 2.459293 | lrm: 0.46 | dt: 1190.32ms | tok/sec: 880,916 | bf16_mfu: 54.11 | total time: 79.48m | eta: 31.3m step 03998/05568 (71.80%) | loss: 2.458150 | lrm: 0.46 | dt: 1196.64ms | tok/sec: 876,265 | bf16_mfu: 53.83 | total time: 79.50m | eta: 31.3m step 03999/05568 (71.82%) | loss: 2.454989 | lrm: 0.46 | dt: 1193.69ms | tok/sec: 878,431 | bf16_mfu: 53.96 | total time: 79.52m | eta: 31.3m step:4000/5568 val_bpb:0.747155 val_time:3.90s step 04000/05568 (71.84%) | loss: 2.454715 | lrm: 0.46 | dt: 1195.45ms | tok/sec: 877,142 | bf16_mfu: 53.88 | total time: 79.54m | eta: 31.2m step 04001/05568 (71.86%) | loss: 2.454804 | lrm: 0.46 | dt: 1194.66ms | tok/sec: 877,716 | bf16_mfu: 53.92 | total time: 79.56m | eta: 31.2m step 04002/05568 (71.88%) | loss: 2.452272 | lrm: 0.46 | dt: 1191.18ms | tok/sec: 880,283 | bf16_mfu: 54.07 | total time: 79.58m | eta: 31.2m step 04003/05568 (71.89%) | loss: 2.439311 | lrm: 0.46 | dt: 1196.26ms | tok/sec: 876,544 | bf16_mfu: 53.84 | total time: 79.60m | eta: 31.2m step 04004/05568 (71.91%) | loss: 2.438035 | lrm: 0.46 | dt: 1195.94ms | tok/sec: 876,781 | bf16_mfu: 53.86 | total time: 79.62m | eta: 31.2m step 04005/05568 (71.93%) | loss: 2.439961 | lrm: 0.46 | dt: 1191.17ms | tok/sec: 880,287 | bf16_mfu: 54.07 | total time: 79.64m | eta: 31.1m step 04006/05568 (71.95%) | loss: 2.438036 | lrm: 0.46 | dt: 1197.00ms | tok/sec: 876,005 | bf16_mfu: 53.81 | total time: 79.65m | eta: 31.1m step 04007/05568 (71.96%) | loss: 2.435999 | lrm: 0.46 | dt: 1195.05ms | tok/sec: 877,433 | bf16_mfu: 53.90 | total time: 79.67m | eta: 31.1m step 04008/05568 (71.98%) | loss: 2.428423 | lrm: 0.46 | dt: 1194.09ms | tok/sec: 878,140 | bf16_mfu: 53.94 | total time: 79.69m | eta: 31.1m step 04009/05568 (72.00%) | loss: 2.430119 | lrm: 0.46 | dt: 1192.74ms | tok/sec: 879,133 | bf16_mfu: 54.00 | total time: 79.71m | eta: 31.1m step 04010/05568 (72.02%) | loss: 2.451306 | lrm: 0.46 | dt: 1193.40ms | tok/sec: 878,646 | bf16_mfu: 53.97 | total time: 79.73m | eta: 31.0m step 04011/05568 (72.04%) | loss: 2.451829 | lrm: 0.46 | dt: 1190.65ms | tok/sec: 880,678 | bf16_mfu: 54.10 | total time: 79.75m | eta: 31.0m step 04012/05568 (72.05%) | loss: 2.451148 | lrm: 0.46 | dt: 1195.15ms | tok/sec: 877,360 | bf16_mfu: 53.89 | total time: 79.77m | eta: 31.0m step 04013/05568 (72.07%) | loss: 2.443234 | lrm: 0.46 | dt: 1195.86ms | tok/sec: 876,838 | bf16_mfu: 53.86 | total time: 79.79m | eta: 31.0m step 04014/05568 (72.09%) | loss: 2.451338 | lrm: 0.46 | dt: 1194.20ms | tok/sec: 878,060 | bf16_mfu: 53.94 | total time: 79.81m | eta: 31.0m step 04015/05568 (72.11%) | loss: 2.443179 | lrm: 0.46 | dt: 1190.89ms | tok/sec: 880,497 | bf16_mfu: 54.09 | total time: 79.83m | eta: 30.9m step 04016/05568 (72.13%) | loss: 2.448296 | lrm: 0.46 | dt: 1191.29ms | tok/sec: 880,201 | bf16_mfu: 54.07 | total time: 79.85m | eta: 30.9m step 04017/05568 (72.14%) | loss: 2.459958 | lrm: 0.46 | dt: 1199.80ms | tok/sec: 873,960 | bf16_mfu: 53.69 | total time: 79.87m | eta: 30.9m step 04018/05568 (72.16%) | loss: 2.468712 | lrm: 0.46 | dt: 1194.19ms | tok/sec: 878,064 | bf16_mfu: 53.94 | total time: 79.89m | eta: 30.9m step 04019/05568 (72.18%) | loss: 2.471130 | lrm: 0.46 | dt: 1197.72ms | tok/sec: 875,478 | bf16_mfu: 53.78 | total time: 79.91m | eta: 30.9m step 04020/05568 (72.20%) | loss: 2.469384 | lrm: 0.46 | dt: 1197.82ms | tok/sec: 875,401 | bf16_mfu: 53.77 | total time: 79.93m | eta: 30.8m step 04021/05568 (72.22%) | loss: 2.470912 | lrm: 0.46 | dt: 1194.43ms | tok/sec: 877,888 | bf16_mfu: 53.93 | total time: 79.95m | eta: 30.8m step 04022/05568 (72.23%) | loss: 2.472083 | lrm: 0.46 | dt: 1194.62ms | tok/sec: 877,745 | bf16_mfu: 53.92 | total time: 79.97m | eta: 30.8m step 04023/05568 (72.25%) | loss: 2.472805 | lrm: 0.46 | dt: 1193.93ms | tok/sec: 878,259 | bf16_mfu: 53.95 | total time: 79.99m | eta: 30.8m step 04024/05568 (72.27%) | loss: 2.466516 | lrm: 0.46 | dt: 1196.12ms | tok/sec: 876,645 | bf16_mfu: 53.85 | total time: 80.01m | eta: 30.8m step 04025/05568 (72.29%) | loss: 2.467414 | lrm: 0.46 | dt: 1193.79ms | tok/sec: 878,360 | bf16_mfu: 53.96 | total time: 80.03m | eta: 30.7m step 04026/05568 (72.31%) | loss: 2.462029 | lrm: 0.45 | dt: 1193.20ms | tok/sec: 878,793 | bf16_mfu: 53.98 | total time: 80.05m | eta: 30.7m step 04027/05568 (72.32%) | loss: 2.451909 | lrm: 0.45 | dt: 1196.91ms | tok/sec: 876,066 | bf16_mfu: 53.81 | total time: 80.07m | eta: 30.7m step 04028/05568 (72.34%) | loss: 2.451462 | lrm: 0.45 | dt: 1195.38ms | tok/sec: 877,193 | bf16_mfu: 53.88 | total time: 80.09m | eta: 30.7m step 04029/05568 (72.36%) | loss: 2.457800 | lrm: 0.45 | dt: 1195.72ms | tok/sec: 876,942 | bf16_mfu: 53.87 | total time: 80.11m | eta: 30.7m step 04030/05568 (72.38%) | loss: 2.462563 | lrm: 0.45 | dt: 1197.51ms | tok/sec: 875,630 | bf16_mfu: 53.79 | total time: 80.13m | eta: 30.6m step 04031/05568 (72.40%) | loss: 2.462964 | lrm: 0.45 | dt: 1200.23ms | tok/sec: 873,645 | bf16_mfu: 53.67 | total time: 80.15m | eta: 30.6m step 04032/05568 (72.41%) | loss: 2.458765 | lrm: 0.45 | dt: 1194.67ms | tok/sec: 877,713 | bf16_mfu: 53.92 | total time: 80.17m | eta: 30.6m step 04033/05568 (72.43%) | loss: 2.458002 | lrm: 0.45 | dt: 1198.36ms | tok/sec: 875,009 | bf16_mfu: 53.75 | total time: 80.19m | eta: 30.6m step 04034/05568 (72.45%) | loss: 2.450812 | lrm: 0.45 | dt: 1194.08ms | tok/sec: 878,148 | bf16_mfu: 53.94 | total time: 80.21m | eta: 30.6m step 04035/05568 (72.47%) | loss: 2.439402 | lrm: 0.45 | dt: 1193.12ms | tok/sec: 878,853 | bf16_mfu: 53.99 | total time: 80.23m | eta: 30.5m step 04036/05568 (72.49%) | loss: 2.449128 | lrm: 0.45 | dt: 1195.68ms | tok/sec: 876,970 | bf16_mfu: 53.87 | total time: 80.25m | eta: 30.5m step 04037/05568 (72.50%) | loss: 2.442702 | lrm: 0.45 | dt: 1196.60ms | tok/sec: 876,296 | bf16_mfu: 53.83 | total time: 80.27m | eta: 30.5m step 04038/05568 (72.52%) | loss: 2.454909 | lrm: 0.45 | dt: 1199.58ms | tok/sec: 874,120 | bf16_mfu: 53.70 | total time: 80.29m | eta: 30.5m step 04039/05568 (72.54%) | loss: 2.454371 | lrm: 0.45 | dt: 1190.52ms | tok/sec: 880,769 | bf16_mfu: 54.10 | total time: 80.31m | eta: 30.5m step 04040/05568 (72.56%) | loss: 2.459383 | lrm: 0.45 | dt: 1195.22ms | tok/sec: 877,308 | bf16_mfu: 53.89 | total time: 80.33m | eta: 30.4m step 04041/05568 (72.58%) | loss: 2.468280 | lrm: 0.45 | dt: 1193.50ms | tok/sec: 878,568 | bf16_mfu: 53.97 | total time: 80.35m | eta: 30.4m step 04042/05568 (72.59%) | loss: 2.471887 | lrm: 0.45 | dt: 1192.04ms | tok/sec: 879,650 | bf16_mfu: 54.03 | total time: 80.37m | eta: 30.4m step 04043/05568 (72.61%) | loss: 2.464337 | lrm: 0.45 | dt: 1197.37ms | tok/sec: 875,734 | bf16_mfu: 53.79 | total time: 80.39m | eta: 30.4m step 04044/05568 (72.63%) | loss: 2.455716 | lrm: 0.45 | dt: 1196.65ms | tok/sec: 876,256 | bf16_mfu: 53.83 | total time: 80.41m | eta: 30.4m step 04045/05568 (72.65%) | loss: 2.458108 | lrm: 0.45 | dt: 1194.52ms | tok/sec: 877,820 | bf16_mfu: 53.92 | total time: 80.43m | eta: 30.3m step 04046/05568 (72.67%) | loss: 2.453338 | lrm: 0.45 | dt: 1195.34ms | tok/sec: 877,219 | bf16_mfu: 53.89 | total time: 80.45m | eta: 30.3m step 04047/05568 (72.68%) | loss: 2.445108 | lrm: 0.45 | dt: 1192.62ms | tok/sec: 879,223 | bf16_mfu: 54.01 | total time: 80.47m | eta: 30.3m step 04048/05568 (72.70%) | loss: 2.443048 | lrm: 0.45 | dt: 1193.12ms | tok/sec: 878,852 | bf16_mfu: 53.99 | total time: 80.49m | eta: 30.3m step 04049/05568 (72.72%) | loss: 2.451099 | lrm: 0.45 | dt: 1196.19ms | tok/sec: 876,593 | bf16_mfu: 53.85 | total time: 80.51m | eta: 30.3m step 04050/05568 (72.74%) | loss: 2.447270 | lrm: 0.45 | dt: 1193.81ms | tok/sec: 878,345 | bf16_mfu: 53.95 | total time: 80.53m | eta: 30.2m step 04051/05568 (72.76%) | loss: 2.450219 | lrm: 0.45 | dt: 1196.20ms | tok/sec: 876,588 | bf16_mfu: 53.85 | total time: 80.55m | eta: 30.2m step 04052/05568 (72.77%) | loss: 2.445964 | lrm: 0.45 | dt: 1194.82ms | tok/sec: 877,601 | bf16_mfu: 53.91 | total time: 80.57m | eta: 30.2m step 04053/05568 (72.79%) | loss: 2.445584 | lrm: 0.45 | dt: 1190.00ms | tok/sec: 881,153 | bf16_mfu: 54.13 | total time: 80.59m | eta: 30.2m step 04054/05568 (72.81%) | loss: 2.439997 | lrm: 0.45 | dt: 1195.75ms | tok/sec: 876,921 | bf16_mfu: 53.87 | total time: 80.61m | eta: 30.2m step 04055/05568 (72.83%) | loss: 2.441534 | lrm: 0.45 | dt: 1197.70ms | tok/sec: 875,492 | bf16_mfu: 53.78 | total time: 80.63m | eta: 30.1m step 04056/05568 (72.84%) | loss: 2.447092 | lrm: 0.45 | dt: 1195.27ms | tok/sec: 877,271 | bf16_mfu: 53.89 | total time: 80.65m | eta: 30.1m step 04057/05568 (72.86%) | loss: 2.442237 | lrm: 0.45 | dt: 1199.68ms | tok/sec: 874,043 | bf16_mfu: 53.69 | total time: 80.67m | eta: 30.1m step 04058/05568 (72.88%) | loss: 2.437228 | lrm: 0.45 | dt: 1194.23ms | tok/sec: 878,034 | bf16_mfu: 53.94 | total time: 80.69m | eta: 30.1m step 04059/05568 (72.90%) | loss: 2.460157 | lrm: 0.45 | dt: 1193.60ms | tok/sec: 878,496 | bf16_mfu: 53.96 | total time: 80.71m | eta: 30.1m step 04060/05568 (72.92%) | loss: 2.467108 | lrm: 0.45 | dt: 1196.31ms | tok/sec: 876,511 | bf16_mfu: 53.84 | total time: 80.73m | eta: 30.0m step 04061/05568 (72.93%) | loss: 2.474773 | lrm: 0.45 | dt: 1194.55ms | tok/sec: 877,802 | bf16_mfu: 53.92 | total time: 80.75m | eta: 30.0m step 04062/05568 (72.95%) | loss: 2.464491 | lrm: 0.45 | dt: 1194.65ms | tok/sec: 877,730 | bf16_mfu: 53.92 | total time: 80.77m | eta: 30.0m step 04063/05568 (72.97%) | loss: 2.463406 | lrm: 0.45 | dt: 1197.59ms | tok/sec: 875,574 | bf16_mfu: 53.78 | total time: 80.79m | eta: 30.0m step 04064/05568 (72.99%) | loss: 2.461287 | lrm: 0.44 | dt: 1195.91ms | tok/sec: 876,804 | bf16_mfu: 53.86 | total time: 80.81m | eta: 30.0m step 04065/05568 (73.01%) | loss: 2.448423 | lrm: 0.44 | dt: 1191.70ms | tok/sec: 879,896 | bf16_mfu: 54.05 | total time: 80.83m | eta: 29.9m step 04066/05568 (73.02%) | loss: 2.453445 | lrm: 0.44 | dt: 1193.78ms | tok/sec: 878,365 | bf16_mfu: 53.96 | total time: 80.85m | eta: 29.9m step 04067/05568 (73.04%) | loss: 2.463726 | lrm: 0.44 | dt: 1196.19ms | tok/sec: 876,594 | bf16_mfu: 53.85 | total time: 80.87m | eta: 29.9m step 04068/05568 (73.06%) | loss: 2.459300 | lrm: 0.44 | dt: 1196.36ms | tok/sec: 876,473 | bf16_mfu: 53.84 | total time: 80.89m | eta: 29.9m step 04069/05568 (73.08%) | loss: 2.452310 | lrm: 0.44 | dt: 1194.94ms | tok/sec: 877,516 | bf16_mfu: 53.90 | total time: 80.91m | eta: 29.9m step 04070/05568 (73.10%) | loss: 2.453148 | lrm: 0.44 | dt: 1197.25ms | tok/sec: 875,816 | bf16_mfu: 53.80 | total time: 80.93m | eta: 29.8m step 04071/05568 (73.11%) | loss: 2.461984 | lrm: 0.44 | dt: 1194.51ms | tok/sec: 877,831 | bf16_mfu: 53.92 | total time: 80.95m | eta: 29.8m step 04072/05568 (73.13%) | loss: 2.462421 | lrm: 0.44 | dt: 1198.54ms | tok/sec: 874,875 | bf16_mfu: 53.74 | total time: 80.97m | eta: 29.8m step 04073/05568 (73.15%) | loss: 2.458633 | lrm: 0.44 | dt: 1194.81ms | tok/sec: 877,607 | bf16_mfu: 53.91 | total time: 80.99m | eta: 29.8m step 04074/05568 (73.17%) | loss: 2.464126 | lrm: 0.44 | dt: 1193.40ms | tok/sec: 878,649 | bf16_mfu: 53.97 | total time: 81.01m | eta: 29.8m step 04075/05568 (73.19%) | loss: 2.462504 | lrm: 0.44 | dt: 1189.61ms | tok/sec: 881,444 | bf16_mfu: 54.15 | total time: 81.03m | eta: 29.7m step 04076/05568 (73.20%) | loss: 2.458583 | lrm: 0.44 | dt: 1198.04ms | tok/sec: 875,246 | bf16_mfu: 53.76 | total time: 81.05m | eta: 29.7m step 04077/05568 (73.22%) | loss: 2.462863 | lrm: 0.44 | dt: 1194.53ms | tok/sec: 877,812 | bf16_mfu: 53.92 | total time: 81.07m | eta: 29.7m step 04078/05568 (73.24%) | loss: 2.460502 | lrm: 0.44 | dt: 1197.51ms | tok/sec: 875,632 | bf16_mfu: 53.79 | total time: 81.09m | eta: 29.7m step 04079/05568 (73.26%) | loss: 2.450826 | lrm: 0.44 | dt: 1194.15ms | tok/sec: 878,093 | bf16_mfu: 53.94 | total time: 81.11m | eta: 29.7m step 04080/05568 (73.28%) | loss: 2.453397 | lrm: 0.44 | dt: 1195.33ms | tok/sec: 877,228 | bf16_mfu: 53.89 | total time: 81.13m | eta: 29.6m step 04081/05568 (73.29%) | loss: 2.450557 | lrm: 0.44 | dt: 1194.62ms | tok/sec: 877,751 | bf16_mfu: 53.92 | total time: 81.15m | eta: 29.6m step 04082/05568 (73.31%) | loss: 2.453718 | lrm: 0.44 | dt: 1193.47ms | tok/sec: 878,597 | bf16_mfu: 53.97 | total time: 81.17m | eta: 29.6m step 04083/05568 (73.33%) | loss: 2.454921 | lrm: 0.44 | dt: 1197.97ms | tok/sec: 875,295 | bf16_mfu: 53.77 | total time: 81.19m | eta: 29.6m step 04084/05568 (73.35%) | loss: 2.454739 | lrm: 0.44 | dt: 1191.30ms | tok/sec: 880,196 | bf16_mfu: 54.07 | total time: 81.21m | eta: 29.6m step 04085/05568 (73.37%) | loss: 2.460671 | lrm: 0.44 | dt: 1199.03ms | tok/sec: 874,522 | bf16_mfu: 53.72 | total time: 81.23m | eta: 29.5m step 04086/05568 (73.38%) | loss: 2.464615 | lrm: 0.44 | dt: 1197.13ms | tok/sec: 875,905 | bf16_mfu: 53.80 | total time: 81.25m | eta: 29.5m step 04087/05568 (73.40%) | loss: 2.457107 | lrm: 0.44 | dt: 1192.70ms | tok/sec: 879,160 | bf16_mfu: 54.00 | total time: 81.27m | eta: 29.5m step 04088/05568 (73.42%) | loss: 2.457723 | lrm: 0.44 | dt: 1196.40ms | tok/sec: 876,445 | bf16_mfu: 53.84 | total time: 81.29m | eta: 29.5m step 04089/05568 (73.44%) | loss: 2.458232 | lrm: 0.44 | dt: 1194.66ms | tok/sec: 877,716 | bf16_mfu: 53.92 | total time: 81.31m | eta: 29.5m step 04090/05568 (73.46%) | loss: 2.464981 | lrm: 0.44 | dt: 1195.02ms | tok/sec: 877,457 | bf16_mfu: 53.90 | total time: 81.33m | eta: 29.4m step 04091/05568 (73.47%) | loss: 2.466660 | lrm: 0.44 | dt: 1195.84ms | tok/sec: 876,851 | bf16_mfu: 53.86 | total time: 81.35m | eta: 29.4m step 04092/05568 (73.49%) | loss: 2.468946 | lrm: 0.44 | dt: 1191.83ms | tok/sec: 879,801 | bf16_mfu: 54.04 | total time: 81.37m | eta: 29.4m step 04093/05568 (73.51%) | loss: 2.476522 | lrm: 0.44 | dt: 1197.46ms | tok/sec: 875,669 | bf16_mfu: 53.79 | total time: 81.39m | eta: 29.4m step 04094/05568 (73.53%) | loss: 2.467475 | lrm: 0.44 | dt: 1192.62ms | tok/sec: 879,219 | bf16_mfu: 54.01 | total time: 81.41m | eta: 29.4m step 04095/05568 (73.55%) | loss: 2.473258 | lrm: 0.44 | dt: 1196.24ms | tok/sec: 876,562 | bf16_mfu: 53.85 | total time: 81.43m | eta: 29.3m step 04096/05568 (73.56%) | loss: 2.468308 | lrm: 0.44 | dt: 1193.62ms | tok/sec: 878,486 | bf16_mfu: 53.96 | total time: 81.45m | eta: 29.3m step 04097/05568 (73.58%) | loss: 2.465237 | lrm: 0.44 | dt: 1195.02ms | tok/sec: 877,455 | bf16_mfu: 53.90 | total time: 81.47m | eta: 29.3m step 04098/05568 (73.60%) | loss: 2.462673 | lrm: 0.44 | dt: 1199.53ms | tok/sec: 874,158 | bf16_mfu: 53.70 | total time: 81.49m | eta: 29.3m step 04099/05568 (73.62%) | loss: 2.464086 | lrm: 0.44 | dt: 1194.39ms | tok/sec: 877,915 | bf16_mfu: 53.93 | total time: 81.51m | eta: 29.3m step 04100/05568 (73.64%) | loss: 2.461420 | lrm: 0.44 | dt: 1192.44ms | tok/sec: 879,349 | bf16_mfu: 54.02 | total time: 81.53m | eta: 29.2m step 04101/05568 (73.65%) | loss: 2.461049 | lrm: 0.44 | dt: 1197.22ms | tok/sec: 875,838 | bf16_mfu: 53.80 | total time: 81.55m | eta: 29.2m step 04102/05568 (73.67%) | loss: 2.467142 | lrm: 0.43 | dt: 1194.62ms | tok/sec: 877,748 | bf16_mfu: 53.92 | total time: 81.57m | eta: 29.2m step 04103/05568 (73.69%) | loss: 2.457574 | lrm: 0.43 | dt: 1192.59ms | tok/sec: 879,243 | bf16_mfu: 54.01 | total time: 81.59m | eta: 29.2m step 04104/05568 (73.71%) | loss: 2.465046 | lrm: 0.43 | dt: 1190.19ms | tok/sec: 881,018 | bf16_mfu: 54.12 | total time: 81.61m | eta: 29.2m step 04105/05568 (73.72%) | loss: 2.472880 | lrm: 0.43 | dt: 1195.51ms | tok/sec: 877,095 | bf16_mfu: 53.88 | total time: 81.63m | eta: 29.1m step 04106/05568 (73.74%) | loss: 2.479988 | lrm: 0.43 | dt: 1195.10ms | tok/sec: 877,392 | bf16_mfu: 53.90 | total time: 81.65m | eta: 29.1m step 04107/05568 (73.76%) | loss: 2.473839 | lrm: 0.43 | dt: 1197.42ms | tok/sec: 875,693 | bf16_mfu: 53.79 | total time: 81.67m | eta: 29.1m step 04108/05568 (73.78%) | loss: 2.478102 | lrm: 0.43 | dt: 1194.50ms | tok/sec: 877,836 | bf16_mfu: 53.92 | total time: 81.69m | eta: 29.1m step 04109/05568 (73.80%) | loss: 2.471213 | lrm: 0.43 | dt: 1196.08ms | tok/sec: 876,676 | bf16_mfu: 53.85 | total time: 81.71m | eta: 29.1m step 04110/05568 (73.81%) | loss: 2.457382 | lrm: 0.43 | dt: 1192.56ms | tok/sec: 879,267 | bf16_mfu: 54.01 | total time: 81.73m | eta: 29.0m step 04111/05568 (73.83%) | loss: 2.454840 | lrm: 0.43 | dt: 1193.05ms | tok/sec: 878,905 | bf16_mfu: 53.99 | total time: 81.75m | eta: 29.0m step 04112/05568 (73.85%) | loss: 2.453604 | lrm: 0.43 | dt: 1193.17ms | tok/sec: 878,815 | bf16_mfu: 53.98 | total time: 81.77m | eta: 29.0m step 04113/05568 (73.87%) | loss: 2.446353 | lrm: 0.43 | dt: 1193.08ms | tok/sec: 878,883 | bf16_mfu: 53.99 | total time: 81.79m | eta: 29.0m step 04114/05568 (73.89%) | loss: 2.438197 | lrm: 0.43 | dt: 1194.02ms | tok/sec: 878,190 | bf16_mfu: 53.95 | total time: 81.81m | eta: 29.0m step 04115/05568 (73.90%) | loss: 2.441359 | lrm: 0.43 | dt: 1195.35ms | tok/sec: 877,214 | bf16_mfu: 53.89 | total time: 81.83m | eta: 28.9m step 04116/05568 (73.92%) | loss: 2.435886 | lrm: 0.43 | dt: 1193.76ms | tok/sec: 878,379 | bf16_mfu: 53.96 | total time: 81.85m | eta: 28.9m step 04117/05568 (73.94%) | loss: 2.443889 | lrm: 0.43 | dt: 1196.03ms | tok/sec: 876,715 | bf16_mfu: 53.85 | total time: 81.87m | eta: 28.9m step 04118/05568 (73.96%) | loss: 2.454264 | lrm: 0.43 | dt: 1195.88ms | tok/sec: 876,821 | bf16_mfu: 53.86 | total time: 81.89m | eta: 28.9m step 04119/05568 (73.98%) | loss: 2.463826 | lrm: 0.43 | dt: 1197.39ms | tok/sec: 875,716 | bf16_mfu: 53.79 | total time: 81.91m | eta: 28.9m step 04120/05568 (73.99%) | loss: 2.467894 | lrm: 0.43 | dt: 1198.17ms | tok/sec: 875,151 | bf16_mfu: 53.76 | total time: 81.93m | eta: 28.8m step 04121/05568 (74.01%) | loss: 2.474703 | lrm: 0.43 | dt: 1192.87ms | tok/sec: 879,037 | bf16_mfu: 54.00 | total time: 81.95m | eta: 28.8m step 04122/05568 (74.03%) | loss: 2.479332 | lrm: 0.43 | dt: 1194.21ms | tok/sec: 878,047 | bf16_mfu: 53.94 | total time: 81.97m | eta: 28.8m step 04123/05568 (74.05%) | loss: 2.474418 | lrm: 0.43 | dt: 1192.76ms | tok/sec: 879,118 | bf16_mfu: 54.00 | total time: 81.99m | eta: 28.8m step 04124/05568 (74.07%) | loss: 2.465454 | lrm: 0.43 | dt: 1197.67ms | tok/sec: 875,516 | bf16_mfu: 53.78 | total time: 82.01m | eta: 28.8m step 04125/05568 (74.08%) | loss: 2.471620 | lrm: 0.43 | dt: 1190.43ms | tok/sec: 880,839 | bf16_mfu: 54.11 | total time: 82.02m | eta: 28.7m step 04126/05568 (74.10%) | loss: 2.462648 | lrm: 0.43 | dt: 1196.07ms | tok/sec: 876,682 | bf16_mfu: 53.85 | total time: 82.04m | eta: 28.7m step 04127/05568 (74.12%) | loss: 2.460022 | lrm: 0.43 | dt: 1195.80ms | tok/sec: 876,884 | bf16_mfu: 53.87 | total time: 82.06m | eta: 28.7m step 04128/05568 (74.14%) | loss: 2.449111 | lrm: 0.43 | dt: 1194.80ms | tok/sec: 877,617 | bf16_mfu: 53.91 | total time: 82.08m | eta: 28.7m step 04129/05568 (74.16%) | loss: 2.450225 | lrm: 0.43 | dt: 1196.02ms | tok/sec: 876,721 | bf16_mfu: 53.85 | total time: 82.10m | eta: 28.7m step 04130/05568 (74.17%) | loss: 2.456510 | lrm: 0.43 | dt: 1191.75ms | tok/sec: 879,865 | bf16_mfu: 54.05 | total time: 82.12m | eta: 28.6m step 04131/05568 (74.19%) | loss: 2.462258 | lrm: 0.43 | dt: 1195.01ms | tok/sec: 877,465 | bf16_mfu: 53.90 | total time: 82.14m | eta: 28.6m step 04132/05568 (74.21%) | loss: 2.460044 | lrm: 0.43 | dt: 1196.72ms | tok/sec: 876,211 | bf16_mfu: 53.82 | total time: 82.16m | eta: 28.6m step 04133/05568 (74.23%) | loss: 2.458572 | lrm: 0.43 | dt: 1197.21ms | tok/sec: 875,848 | bf16_mfu: 53.80 | total time: 82.18m | eta: 28.6m step 04134/05568 (74.25%) | loss: 2.453981 | lrm: 0.43 | dt: 1197.21ms | tok/sec: 875,847 | bf16_mfu: 53.80 | total time: 82.20m | eta: 28.6m step 04135/05568 (74.26%) | loss: 2.449547 | lrm: 0.43 | dt: 1194.38ms | tok/sec: 877,926 | bf16_mfu: 53.93 | total time: 82.22m | eta: 28.5m step 04136/05568 (74.28%) | loss: 2.455652 | lrm: 0.43 | dt: 1191.12ms | tok/sec: 880,327 | bf16_mfu: 54.08 | total time: 82.24m | eta: 28.5m step 04137/05568 (74.30%) | loss: 2.464068 | lrm: 0.43 | dt: 1189.31ms | tok/sec: 881,671 | bf16_mfu: 54.16 | total time: 82.26m | eta: 28.5m step 04138/05568 (74.32%) | loss: 2.458693 | lrm: 0.43 | dt: 1193.65ms | tok/sec: 878,463 | bf16_mfu: 53.96 | total time: 82.28m | eta: 28.5m step 04139/05568 (74.34%) | loss: 2.452633 | lrm: 0.43 | dt: 1194.74ms | tok/sec: 877,663 | bf16_mfu: 53.91 | total time: 82.30m | eta: 28.5m step 04140/05568 (74.35%) | loss: 2.447459 | lrm: 0.42 | dt: 1195.84ms | tok/sec: 876,855 | bf16_mfu: 53.86 | total time: 82.32m | eta: 28.4m step 04141/05568 (74.37%) | loss: 2.455982 | lrm: 0.42 | dt: 1191.49ms | tok/sec: 880,055 | bf16_mfu: 54.06 | total time: 82.34m | eta: 28.4m step 04142/05568 (74.39%) | loss: 2.460322 | lrm: 0.42 | dt: 1197.23ms | tok/sec: 875,831 | bf16_mfu: 53.80 | total time: 82.36m | eta: 28.4m step 04143/05568 (74.41%) | loss: 2.455612 | lrm: 0.42 | dt: 1195.59ms | tok/sec: 877,034 | bf16_mfu: 53.87 | total time: 82.38m | eta: 28.4m step 04144/05568 (74.43%) | loss: 2.452222 | lrm: 0.42 | dt: 1190.70ms | tok/sec: 880,640 | bf16_mfu: 54.10 | total time: 82.40m | eta: 28.4m step 04145/05568 (74.44%) | loss: 2.455324 | lrm: 0.42 | dt: 1196.07ms | tok/sec: 876,687 | bf16_mfu: 53.85 | total time: 82.42m | eta: 28.3m step 04146/05568 (74.46%) | loss: 2.459793 | lrm: 0.42 | dt: 1195.36ms | tok/sec: 877,203 | bf16_mfu: 53.88 | total time: 82.44m | eta: 28.3m step 04147/05568 (74.48%) | loss: 2.449597 | lrm: 0.42 | dt: 1194.13ms | tok/sec: 878,109 | bf16_mfu: 53.94 | total time: 82.46m | eta: 28.3m step 04148/05568 (74.50%) | loss: 2.444504 | lrm: 0.42 | dt: 1195.76ms | tok/sec: 876,910 | bf16_mfu: 53.87 | total time: 82.48m | eta: 28.3m step 04149/05568 (74.52%) | loss: 2.445889 | lrm: 0.42 | dt: 1194.85ms | tok/sec: 877,576 | bf16_mfu: 53.91 | total time: 82.50m | eta: 28.3m step 04150/05568 (74.53%) | loss: 2.449946 | lrm: 0.42 | dt: 1195.22ms | tok/sec: 877,305 | bf16_mfu: 53.89 | total time: 82.52m | eta: 28.2m step 04151/05568 (74.55%) | loss: 2.434356 | lrm: 0.42 | dt: 1195.44ms | tok/sec: 877,144 | bf16_mfu: 53.88 | total time: 82.54m | eta: 28.2m step 04152/05568 (74.57%) | loss: 2.430638 | lrm: 0.42 | dt: 1193.07ms | tok/sec: 878,886 | bf16_mfu: 53.99 | total time: 82.56m | eta: 28.2m step 04153/05568 (74.59%) | loss: 2.439298 | lrm: 0.42 | dt: 1195.86ms | tok/sec: 876,842 | bf16_mfu: 53.86 | total time: 82.58m | eta: 28.2m step 04154/05568 (74.60%) | loss: 2.439271 | lrm: 0.42 | dt: 1194.43ms | tok/sec: 877,887 | bf16_mfu: 53.93 | total time: 82.60m | eta: 28.2m step 04155/05568 (74.62%) | loss: 2.437272 | lrm: 0.42 | dt: 1193.77ms | tok/sec: 878,371 | bf16_mfu: 53.96 | total time: 82.62m | eta: 28.1m step 04156/05568 (74.64%) | loss: 2.441193 | lrm: 0.42 | dt: 1196.04ms | tok/sec: 876,707 | bf16_mfu: 53.85 | total time: 82.64m | eta: 28.1m step 04157/05568 (74.66%) | loss: 2.438132 | lrm: 0.42 | dt: 1197.27ms | tok/sec: 875,809 | bf16_mfu: 53.80 | total time: 82.66m | eta: 28.1m step 04158/05568 (74.68%) | loss: 2.439439 | lrm: 0.42 | dt: 1196.61ms | tok/sec: 876,288 | bf16_mfu: 53.83 | total time: 82.68m | eta: 28.1m step 04159/05568 (74.69%) | loss: 2.442742 | lrm: 0.42 | dt: 1195.41ms | tok/sec: 877,171 | bf16_mfu: 53.88 | total time: 82.70m | eta: 28.1m step 04160/05568 (74.71%) | loss: 2.436771 | lrm: 0.42 | dt: 1194.73ms | tok/sec: 877,666 | bf16_mfu: 53.91 | total time: 82.72m | eta: 28.0m step 04161/05568 (74.73%) | loss: 2.438735 | lrm: 0.42 | dt: 1195.80ms | tok/sec: 876,884 | bf16_mfu: 53.87 | total time: 82.74m | eta: 28.0m step 04162/05568 (74.75%) | loss: 2.430180 | lrm: 0.42 | dt: 1192.53ms | tok/sec: 879,289 | bf16_mfu: 54.01 | total time: 82.76m | eta: 28.0m step 04163/05568 (74.77%) | loss: 2.424508 | lrm: 0.42 | dt: 1195.62ms | tok/sec: 877,016 | bf16_mfu: 53.87 | total time: 82.78m | eta: 28.0m step 04164/05568 (74.78%) | loss: 2.422103 | lrm: 0.42 | dt: 1193.03ms | tok/sec: 878,920 | bf16_mfu: 53.99 | total time: 82.80m | eta: 28.0m step 04165/05568 (74.80%) | loss: 2.398177 | lrm: 0.42 | dt: 1192.84ms | tok/sec: 879,055 | bf16_mfu: 54.00 | total time: 82.82m | eta: 27.9m step 04166/05568 (74.82%) | loss: 2.404079 | lrm: 0.42 | dt: 1191.80ms | tok/sec: 879,823 | bf16_mfu: 54.05 | total time: 82.84m | eta: 27.9m step 04167/05568 (74.84%) | loss: 2.405880 | lrm: 0.42 | dt: 1191.42ms | tok/sec: 880,102 | bf16_mfu: 54.06 | total time: 82.86m | eta: 27.9m step 04168/05568 (74.86%) | loss: 2.423072 | lrm: 0.42 | dt: 1197.08ms | tok/sec: 875,944 | bf16_mfu: 53.81 | total time: 82.88m | eta: 27.9m step 04169/05568 (74.87%) | loss: 2.426838 | lrm: 0.42 | dt: 1193.10ms | tok/sec: 878,866 | bf16_mfu: 53.99 | total time: 82.90m | eta: 27.9m step 04170/05568 (74.89%) | loss: 2.429965 | lrm: 0.42 | dt: 1193.37ms | tok/sec: 878,668 | bf16_mfu: 53.97 | total time: 82.92m | eta: 27.8m step 04171/05568 (74.91%) | loss: 2.422466 | lrm: 0.42 | dt: 1196.90ms | tok/sec: 876,074 | bf16_mfu: 53.82 | total time: 82.94m | eta: 27.8m step 04172/05568 (74.93%) | loss: 2.422948 | lrm: 0.42 | dt: 1193.32ms | tok/sec: 878,707 | bf16_mfu: 53.98 | total time: 82.96m | eta: 27.8m step 04173/05568 (74.95%) | loss: 2.435713 | lrm: 0.42 | dt: 1192.48ms | tok/sec: 879,324 | bf16_mfu: 54.01 | total time: 82.98m | eta: 27.8m step 04174/05568 (74.96%) | loss: 2.428180 | lrm: 0.42 | dt: 1195.30ms | tok/sec: 877,251 | bf16_mfu: 53.89 | total time: 83.00m | eta: 27.8m step 04175/05568 (74.98%) | loss: 2.430458 | lrm: 0.42 | dt: 1194.88ms | tok/sec: 877,558 | bf16_mfu: 53.91 | total time: 83.02m | eta: 27.7m step 04176/05568 (75.00%) | loss: 2.435121 | lrm: 0.42 | dt: 1194.47ms | tok/sec: 877,856 | bf16_mfu: 53.92 | total time: 83.04m | eta: 27.7m step 04177/05568 (75.02%) | loss: 2.443557 | lrm: 0.42 | dt: 1194.02ms | tok/sec: 878,192 | bf16_mfu: 53.95 | total time: 83.06m | eta: 27.7m step 04178/05568 (75.04%) | loss: 2.439608 | lrm: 0.41 | dt: 1193.43ms | tok/sec: 878,624 | bf16_mfu: 53.97 | total time: 83.08m | eta: 27.7m step 04179/05568 (75.05%) | loss: 2.442611 | lrm: 0.41 | dt: 1192.82ms | tok/sec: 879,073 | bf16_mfu: 54.00 | total time: 83.10m | eta: 27.7m step 04180/05568 (75.07%) | loss: 2.440038 | lrm: 0.41 | dt: 1193.59ms | tok/sec: 878,507 | bf16_mfu: 53.96 | total time: 83.12m | eta: 27.6m step 04181/05568 (75.09%) | loss: 2.438510 | lrm: 0.41 | dt: 1196.87ms | tok/sec: 876,095 | bf16_mfu: 53.82 | total time: 83.14m | eta: 27.6m step 04182/05568 (75.11%) | loss: 2.434480 | lrm: 0.41 | dt: 1195.94ms | tok/sec: 876,781 | bf16_mfu: 53.86 | total time: 83.16m | eta: 27.6m step 04183/05568 (75.13%) | loss: 2.437729 | lrm: 0.41 | dt: 1194.18ms | tok/sec: 878,072 | bf16_mfu: 53.94 | total time: 83.18m | eta: 27.6m step 04184/05568 (75.14%) | loss: 2.437042 | lrm: 0.41 | dt: 1195.12ms | tok/sec: 877,383 | bf16_mfu: 53.90 | total time: 83.20m | eta: 27.6m step 04185/05568 (75.16%) | loss: 2.441019 | lrm: 0.41 | dt: 1193.55ms | tok/sec: 878,533 | bf16_mfu: 53.97 | total time: 83.22m | eta: 27.5m step 04186/05568 (75.18%) | loss: 2.447540 | lrm: 0.41 | dt: 1196.16ms | tok/sec: 876,617 | bf16_mfu: 53.85 | total time: 83.24m | eta: 27.5m step 04187/05568 (75.20%) | loss: 2.438868 | lrm: 0.41 | dt: 1197.63ms | tok/sec: 875,544 | bf16_mfu: 53.78 | total time: 83.26m | eta: 27.5m step 04188/05568 (75.22%) | loss: 2.438113 | lrm: 0.41 | dt: 1197.10ms | tok/sec: 875,929 | bf16_mfu: 53.81 | total time: 83.28m | eta: 27.5m step 04189/05568 (75.23%) | loss: 2.444851 | lrm: 0.41 | dt: 1196.55ms | tok/sec: 876,334 | bf16_mfu: 53.83 | total time: 83.30m | eta: 27.5m step 04190/05568 (75.25%) | loss: 2.434745 | lrm: 0.41 | dt: 1195.26ms | tok/sec: 877,277 | bf16_mfu: 53.89 | total time: 83.32m | eta: 27.4m step 04191/05568 (75.27%) | loss: 2.424050 | lrm: 0.41 | dt: 1193.35ms | tok/sec: 878,681 | bf16_mfu: 53.98 | total time: 83.34m | eta: 27.4m step 04192/05568 (75.29%) | loss: 2.419220 | lrm: 0.41 | dt: 1195.18ms | tok/sec: 877,334 | bf16_mfu: 53.89 | total time: 83.36m | eta: 27.4m step 04193/05568 (75.31%) | loss: 2.426278 | lrm: 0.41 | dt: 1190.38ms | tok/sec: 880,872 | bf16_mfu: 54.11 | total time: 83.38m | eta: 27.4m step 04194/05568 (75.32%) | loss: 2.428735 | lrm: 0.41 | dt: 1194.95ms | tok/sec: 877,503 | bf16_mfu: 53.90 | total time: 83.40m | eta: 27.4m step 04195/05568 (75.34%) | loss: 2.434654 | lrm: 0.41 | dt: 1196.02ms | tok/sec: 876,718 | bf16_mfu: 53.85 | total time: 83.42m | eta: 27.3m step 04196/05568 (75.36%) | loss: 2.434516 | lrm: 0.41 | dt: 1199.28ms | tok/sec: 874,339 | bf16_mfu: 53.71 | total time: 83.44m | eta: 27.3m step 04197/05568 (75.38%) | loss: 2.434062 | lrm: 0.41 | dt: 1196.47ms | tok/sec: 876,394 | bf16_mfu: 53.83 | total time: 83.46m | eta: 27.3m step 04198/05568 (75.40%) | loss: 2.443312 | lrm: 0.41 | dt: 1193.70ms | tok/sec: 878,422 | bf16_mfu: 53.96 | total time: 83.48m | eta: 27.3m step 04199/05568 (75.41%) | loss: 2.441664 | lrm: 0.41 | dt: 1199.65ms | tok/sec: 874,070 | bf16_mfu: 53.69 | total time: 83.50m | eta: 27.3m step 04200/05568 (75.43%) | loss: 2.439617 | lrm: 0.41 | dt: 1197.66ms | tok/sec: 875,522 | bf16_mfu: 53.78 | total time: 83.52m | eta: 27.2m step 04201/05568 (75.45%) | loss: 2.447318 | lrm: 0.41 | dt: 1193.60ms | tok/sec: 878,501 | bf16_mfu: 53.96 | total time: 83.54m | eta: 27.2m step 04202/05568 (75.47%) | loss: 2.455302 | lrm: 0.41 | dt: 1192.54ms | tok/sec: 879,279 | bf16_mfu: 54.01 | total time: 83.56m | eta: 27.2m step 04203/05568 (75.48%) | loss: 2.445821 | lrm: 0.41 | dt: 1193.50ms | tok/sec: 878,572 | bf16_mfu: 53.97 | total time: 83.58m | eta: 27.2m step 04204/05568 (75.50%) | loss: 2.436003 | lrm: 0.41 | dt: 1195.67ms | tok/sec: 876,978 | bf16_mfu: 53.87 | total time: 83.60m | eta: 27.2m step 04205/05568 (75.52%) | loss: 2.435025 | lrm: 0.41 | dt: 1191.31ms | tok/sec: 880,184 | bf16_mfu: 54.07 | total time: 83.62m | eta: 27.1m step 04206/05568 (75.54%) | loss: 2.441155 | lrm: 0.41 | dt: 1195.66ms | tok/sec: 876,986 | bf16_mfu: 53.87 | total time: 83.64m | eta: 27.1m step 04207/05568 (75.56%) | loss: 2.445384 | lrm: 0.41 | dt: 1195.57ms | tok/sec: 877,051 | bf16_mfu: 53.88 | total time: 83.66m | eta: 27.1m step 04208/05568 (75.57%) | loss: 2.450222 | lrm: 0.41 | dt: 1192.21ms | tok/sec: 879,525 | bf16_mfu: 54.03 | total time: 83.68m | eta: 27.1m step 04209/05568 (75.59%) | loss: 2.444383 | lrm: 0.41 | dt: 1193.69ms | tok/sec: 878,432 | bf16_mfu: 53.96 | total time: 83.70m | eta: 27.1m step 04210/05568 (75.61%) | loss: 2.444637 | lrm: 0.41 | dt: 1194.73ms | tok/sec: 877,671 | bf16_mfu: 53.91 | total time: 83.72m | eta: 27.0m step 04211/05568 (75.63%) | loss: 2.440735 | lrm: 0.41 | dt: 1189.60ms | tok/sec: 881,451 | bf16_mfu: 54.15 | total time: 83.74m | eta: 27.0m step 04212/05568 (75.65%) | loss: 2.432936 | lrm: 0.41 | dt: 1197.47ms | tok/sec: 875,658 | bf16_mfu: 53.79 | total time: 83.76m | eta: 27.0m step 04213/05568 (75.66%) | loss: 2.432455 | lrm: 0.41 | dt: 1197.90ms | tok/sec: 875,343 | bf16_mfu: 53.77 | total time: 83.78m | eta: 27.0m step 04214/05568 (75.68%) | loss: 2.429770 | lrm: 0.41 | dt: 1196.10ms | tok/sec: 876,664 | bf16_mfu: 53.85 | total time: 83.80m | eta: 27.0m step 04215/05568 (75.70%) | loss: 2.426672 | lrm: 0.41 | dt: 1194.55ms | tok/sec: 877,800 | bf16_mfu: 53.92 | total time: 83.82m | eta: 26.9m step 04216/05568 (75.72%) | loss: 2.431457 | lrm: 0.40 | dt: 1200.95ms | tok/sec: 873,121 | bf16_mfu: 53.63 | total time: 83.84m | eta: 26.9m step 04217/05568 (75.74%) | loss: 2.445063 | lrm: 0.40 | dt: 1199.38ms | tok/sec: 874,261 | bf16_mfu: 53.70 | total time: 83.86m | eta: 26.9m step 04218/05568 (75.75%) | loss: 2.428828 | lrm: 0.40 | dt: 1198.62ms | tok/sec: 874,822 | bf16_mfu: 53.74 | total time: 83.88m | eta: 26.9m step 04219/05568 (75.77%) | loss: 2.416856 | lrm: 0.40 | dt: 1192.33ms | tok/sec: 879,435 | bf16_mfu: 54.02 | total time: 83.90m | eta: 26.9m step 04220/05568 (75.79%) | loss: 2.422568 | lrm: 0.40 | dt: 1195.61ms | tok/sec: 877,024 | bf16_mfu: 53.87 | total time: 83.92m | eta: 26.8m step 04221/05568 (75.81%) | loss: 2.418722 | lrm: 0.40 | dt: 1201.10ms | tok/sec: 873,011 | bf16_mfu: 53.63 | total time: 83.94m | eta: 26.8m step 04222/05568 (75.83%) | loss: 2.427322 | lrm: 0.40 | dt: 1198.40ms | tok/sec: 874,976 | bf16_mfu: 53.75 | total time: 83.96m | eta: 26.8m step 04223/05568 (75.84%) | loss: 2.424505 | lrm: 0.40 | dt: 1196.48ms | tok/sec: 876,387 | bf16_mfu: 53.83 | total time: 83.98m | eta: 26.8m step 04224/05568 (75.86%) | loss: 2.428775 | lrm: 0.40 | dt: 1194.44ms | tok/sec: 877,878 | bf16_mfu: 53.93 | total time: 84.00m | eta: 26.8m step 04225/05568 (75.88%) | loss: 2.426921 | lrm: 0.40 | dt: 1194.30ms | tok/sec: 877,986 | bf16_mfu: 53.93 | total time: 84.02m | eta: 26.7m step 04226/05568 (75.90%) | loss: 2.427419 | lrm: 0.40 | dt: 1194.40ms | tok/sec: 877,906 | bf16_mfu: 53.93 | total time: 84.04m | eta: 26.7m step 04227/05568 (75.92%) | loss: 2.423929 | lrm: 0.40 | dt: 1196.79ms | tok/sec: 876,156 | bf16_mfu: 53.82 | total time: 84.06m | eta: 26.7m step 04228/05568 (75.93%) | loss: 2.430849 | lrm: 0.40 | dt: 1196.44ms | tok/sec: 876,410 | bf16_mfu: 53.84 | total time: 84.08m | eta: 26.7m step 04229/05568 (75.95%) | loss: 2.431252 | lrm: 0.40 | dt: 1191.60ms | tok/sec: 879,976 | bf16_mfu: 54.05 | total time: 84.10m | eta: 26.7m step 04230/05568 (75.97%) | loss: 2.431512 | lrm: 0.40 | dt: 1197.39ms | tok/sec: 875,718 | bf16_mfu: 53.79 | total time: 84.12m | eta: 26.7m step 04231/05568 (75.99%) | loss: 2.431916 | lrm: 0.40 | dt: 1190.99ms | tok/sec: 880,426 | bf16_mfu: 54.08 | total time: 84.14m | eta: 26.6m step 04232/05568 (76.01%) | loss: 2.444946 | lrm: 0.40 | dt: 1201.09ms | tok/sec: 873,018 | bf16_mfu: 53.63 | total time: 84.16m | eta: 26.6m step 04233/05568 (76.02%) | loss: 2.452876 | lrm: 0.40 | dt: 1190.82ms | tok/sec: 880,550 | bf16_mfu: 54.09 | total time: 84.18m | eta: 26.6m step 04234/05568 (76.04%) | loss: 2.445814 | lrm: 0.40 | dt: 1192.04ms | tok/sec: 879,645 | bf16_mfu: 54.03 | total time: 84.20m | eta: 26.6m step 04235/05568 (76.06%) | loss: 2.447848 | lrm: 0.40 | dt: 1195.31ms | tok/sec: 877,244 | bf16_mfu: 53.89 | total time: 84.22m | eta: 26.6m step 04236/05568 (76.08%) | loss: 2.452180 | lrm: 0.40 | dt: 1196.82ms | tok/sec: 876,135 | bf16_mfu: 53.82 | total time: 84.24m | eta: 26.5m step 04237/05568 (76.10%) | loss: 2.462897 | lrm: 0.40 | dt: 1198.02ms | tok/sec: 875,254 | bf16_mfu: 53.76 | total time: 84.26m | eta: 26.5m step 04238/05568 (76.11%) | loss: 2.469788 | lrm: 0.40 | dt: 1194.74ms | tok/sec: 877,660 | bf16_mfu: 53.91 | total time: 84.28m | eta: 26.5m step 04239/05568 (76.13%) | loss: 2.468929 | lrm: 0.40 | dt: 1193.52ms | tok/sec: 878,554 | bf16_mfu: 53.97 | total time: 84.30m | eta: 26.5m step 04240/05568 (76.15%) | loss: 2.468550 | lrm: 0.40 | dt: 1193.38ms | tok/sec: 878,664 | bf16_mfu: 53.97 | total time: 84.32m | eta: 26.5m step 04241/05568 (76.17%) | loss: 2.467248 | lrm: 0.40 | dt: 1194.58ms | tok/sec: 877,776 | bf16_mfu: 53.92 | total time: 84.34m | eta: 26.4m step 04242/05568 (76.19%) | loss: 2.457342 | lrm: 0.40 | dt: 1197.26ms | tok/sec: 875,815 | bf16_mfu: 53.80 | total time: 84.36m | eta: 26.4m step 04243/05568 (76.20%) | loss: 2.449273 | lrm: 0.40 | dt: 1194.62ms | tok/sec: 877,751 | bf16_mfu: 53.92 | total time: 84.38m | eta: 26.4m step 04244/05568 (76.22%) | loss: 2.451789 | lrm: 0.40 | dt: 1192.99ms | tok/sec: 878,946 | bf16_mfu: 53.99 | total time: 84.39m | eta: 26.4m step 04245/05568 (76.24%) | loss: 2.454984 | lrm: 0.40 | dt: 1202.59ms | tok/sec: 871,935 | bf16_mfu: 53.56 | total time: 84.42m | eta: 26.4m step 04246/05568 (76.26%) | loss: 2.454977 | lrm: 0.40 | dt: 1193.72ms | tok/sec: 878,412 | bf16_mfu: 53.96 | total time: 84.43m | eta: 26.3m step 04247/05568 (76.28%) | loss: 2.447124 | lrm: 0.40 | dt: 1191.54ms | tok/sec: 880,019 | bf16_mfu: 54.06 | total time: 84.45m | eta: 26.3m step 04248/05568 (76.29%) | loss: 2.445536 | lrm: 0.40 | dt: 1196.05ms | tok/sec: 876,702 | bf16_mfu: 53.85 | total time: 84.47m | eta: 26.3m step 04249/05568 (76.31%) | loss: 2.442529 | lrm: 0.40 | dt: 1193.94ms | tok/sec: 878,251 | bf16_mfu: 53.95 | total time: 84.49m | eta: 26.3m step:4250/5568 val_bpb:0.741616 val_time:3.90s step 04250/05568 (76.33%) | loss: 2.445399 | lrm: 0.40 | dt: 1195.94ms | tok/sec: 876,782 | bf16_mfu: 53.86 | total time: 84.51m | eta: 26.3m step 04251/05568 (76.35%) | loss: 2.449949 | lrm: 0.40 | dt: 1194.66ms | tok/sec: 877,721 | bf16_mfu: 53.92 | total time: 84.53m | eta: 26.2m step 04252/05568 (76.36%) | loss: 2.438333 | lrm: 0.40 | dt: 1202.65ms | tok/sec: 871,888 | bf16_mfu: 53.56 | total time: 84.55m | eta: 26.2m step 04253/05568 (76.38%) | loss: 2.440161 | lrm: 0.40 | dt: 1193.07ms | tok/sec: 878,889 | bf16_mfu: 53.99 | total time: 84.57m | eta: 26.2m step 04254/05568 (76.40%) | loss: 2.434612 | lrm: 0.39 | dt: 1194.35ms | tok/sec: 877,943 | bf16_mfu: 53.93 | total time: 84.59m | eta: 26.2m step 04255/05568 (76.42%) | loss: 2.437970 | lrm: 0.39 | dt: 1195.18ms | tok/sec: 877,340 | bf16_mfu: 53.89 | total time: 84.61m | eta: 26.2m step 04256/05568 (76.44%) | loss: 2.443385 | lrm: 0.39 | dt: 1196.32ms | tok/sec: 876,498 | bf16_mfu: 53.84 | total time: 84.63m | eta: 26.1m step 04257/05568 (76.45%) | loss: 2.440928 | lrm: 0.39 | dt: 1189.58ms | tok/sec: 881,465 | bf16_mfu: 54.15 | total time: 84.65m | eta: 26.1m step 04258/05568 (76.47%) | loss: 2.429024 | lrm: 0.39 | dt: 1192.50ms | tok/sec: 879,311 | bf16_mfu: 54.01 | total time: 84.67m | eta: 26.1m step 04259/05568 (76.49%) | loss: 2.418922 | lrm: 0.39 | dt: 1194.94ms | tok/sec: 877,514 | bf16_mfu: 53.90 | total time: 84.69m | eta: 26.1m step 04260/05568 (76.51%) | loss: 2.434840 | lrm: 0.39 | dt: 1198.94ms | tok/sec: 874,586 | bf16_mfu: 53.72 | total time: 84.71m | eta: 26.1m step 04261/05568 (76.53%) | loss: 2.442560 | lrm: 0.39 | dt: 1193.92ms | tok/sec: 878,264 | bf16_mfu: 53.95 | total time: 84.73m | eta: 26.0m step 04262/05568 (76.54%) | loss: 2.444374 | lrm: 0.39 | dt: 1196.30ms | tok/sec: 876,518 | bf16_mfu: 53.84 | total time: 84.75m | eta: 26.0m step 04263/05568 (76.56%) | loss: 2.443806 | lrm: 0.39 | dt: 1194.59ms | tok/sec: 877,772 | bf16_mfu: 53.92 | total time: 84.77m | eta: 26.0m step 04264/05568 (76.58%) | loss: 2.442576 | lrm: 0.39 | dt: 1191.80ms | tok/sec: 879,824 | bf16_mfu: 54.05 | total time: 84.79m | eta: 26.0m step 04265/05568 (76.60%) | loss: 2.441116 | lrm: 0.39 | dt: 1198.67ms | tok/sec: 874,779 | bf16_mfu: 53.74 | total time: 84.81m | eta: 26.0m step 04266/05568 (76.62%) | loss: 2.454856 | lrm: 0.39 | dt: 1196.73ms | tok/sec: 876,201 | bf16_mfu: 53.82 | total time: 84.83m | eta: 25.9m step 04267/05568 (76.63%) | loss: 2.454935 | lrm: 0.39 | dt: 1198.87ms | tok/sec: 874,636 | bf16_mfu: 53.73 | total time: 84.85m | eta: 25.9m step 04268/05568 (76.65%) | loss: 2.445190 | lrm: 0.39 | dt: 1201.86ms | tok/sec: 872,462 | bf16_mfu: 53.59 | total time: 84.87m | eta: 25.9m step 04269/05568 (76.67%) | loss: 2.448252 | lrm: 0.39 | dt: 1193.51ms | tok/sec: 878,566 | bf16_mfu: 53.97 | total time: 84.89m | eta: 25.9m step 04270/05568 (76.69%) | loss: 2.463793 | lrm: 0.39 | dt: 1188.77ms | tok/sec: 882,070 | bf16_mfu: 54.18 | total time: 84.91m | eta: 25.9m step 04271/05568 (76.71%) | loss: 2.457842 | lrm: 0.39 | dt: 1197.34ms | tok/sec: 875,758 | bf16_mfu: 53.80 | total time: 84.93m | eta: 25.8m step 04272/05568 (76.72%) | loss: 2.449359 | lrm: 0.39 | dt: 1197.22ms | tok/sec: 875,839 | bf16_mfu: 53.80 | total time: 84.95m | eta: 25.8m step 04273/05568 (76.74%) | loss: 2.455367 | lrm: 0.39 | dt: 1195.11ms | tok/sec: 877,392 | bf16_mfu: 53.90 | total time: 84.97m | eta: 25.8m step 04274/05568 (76.76%) | loss: 2.450787 | lrm: 0.39 | dt: 1195.60ms | tok/sec: 877,027 | bf16_mfu: 53.87 | total time: 84.99m | eta: 25.8m step 04275/05568 (76.78%) | loss: 2.451230 | lrm: 0.39 | dt: 1195.06ms | tok/sec: 877,425 | bf16_mfu: 53.90 | total time: 85.01m | eta: 25.8m step 04276/05568 (76.80%) | loss: 2.458788 | lrm: 0.39 | dt: 1193.18ms | tok/sec: 878,810 | bf16_mfu: 53.98 | total time: 85.03m | eta: 25.7m step 04277/05568 (76.81%) | loss: 2.463161 | lrm: 0.39 | dt: 1197.57ms | tok/sec: 875,584 | bf16_mfu: 53.79 | total time: 85.05m | eta: 25.7m step 04278/05568 (76.83%) | loss: 2.475470 | lrm: 0.39 | dt: 1198.63ms | tok/sec: 874,811 | bf16_mfu: 53.74 | total time: 85.07m | eta: 25.7m step 04279/05568 (76.85%) | loss: 2.470507 | lrm: 0.39 | dt: 1198.43ms | tok/sec: 874,957 | bf16_mfu: 53.75 | total time: 85.09m | eta: 25.7m step 04280/05568 (76.87%) | loss: 2.461857 | lrm: 0.39 | dt: 1191.75ms | tok/sec: 879,864 | bf16_mfu: 54.05 | total time: 85.11m | eta: 25.7m step 04281/05568 (76.89%) | loss: 2.452699 | lrm: 0.39 | dt: 1191.31ms | tok/sec: 880,183 | bf16_mfu: 54.07 | total time: 85.13m | eta: 25.6m step 04282/05568 (76.90%) | loss: 2.451415 | lrm: 0.39 | dt: 1193.16ms | tok/sec: 878,821 | bf16_mfu: 53.98 | total time: 85.15m | eta: 25.6m step 04283/05568 (76.92%) | loss: 2.442673 | lrm: 0.39 | dt: 1190.06ms | tok/sec: 881,108 | bf16_mfu: 54.12 | total time: 85.17m | eta: 25.6m step 04284/05568 (76.94%) | loss: 2.418588 | lrm: 0.39 | dt: 1194.01ms | tok/sec: 878,195 | bf16_mfu: 53.95 | total time: 85.19m | eta: 25.6m step 04285/05568 (76.96%) | loss: 2.419195 | lrm: 0.39 | dt: 1190.62ms | tok/sec: 880,693 | bf16_mfu: 54.10 | total time: 85.21m | eta: 25.6m step 04286/05568 (76.98%) | loss: 2.421922 | lrm: 0.39 | dt: 1193.93ms | tok/sec: 878,254 | bf16_mfu: 53.95 | total time: 85.23m | eta: 25.5m step 04287/05568 (76.99%) | loss: 2.430270 | lrm: 0.39 | dt: 1191.39ms | tok/sec: 880,127 | bf16_mfu: 54.06 | total time: 85.25m | eta: 25.5m step 04288/05568 (77.01%) | loss: 2.429032 | lrm: 0.39 | dt: 1194.05ms | tok/sec: 878,163 | bf16_mfu: 53.94 | total time: 85.27m | eta: 25.5m step 04289/05568 (77.03%) | loss: 2.429608 | lrm: 0.39 | dt: 1196.45ms | tok/sec: 876,404 | bf16_mfu: 53.84 | total time: 85.29m | eta: 25.5m step 04290/05568 (77.05%) | loss: 2.427955 | lrm: 0.39 | dt: 1196.00ms | tok/sec: 876,737 | bf16_mfu: 53.86 | total time: 85.31m | eta: 25.5m step 04291/05568 (77.07%) | loss: 2.441098 | lrm: 0.39 | dt: 1193.73ms | tok/sec: 878,401 | bf16_mfu: 53.96 | total time: 85.33m | eta: 25.4m step 04292/05568 (77.08%) | loss: 2.438941 | lrm: 0.38 | dt: 1197.64ms | tok/sec: 875,535 | bf16_mfu: 53.78 | total time: 85.35m | eta: 25.4m step 04293/05568 (77.10%) | loss: 2.437758 | lrm: 0.38 | dt: 1198.56ms | tok/sec: 874,865 | bf16_mfu: 53.74 | total time: 85.37m | eta: 25.4m step 04294/05568 (77.12%) | loss: 2.435141 | lrm: 0.38 | dt: 1197.32ms | tok/sec: 875,769 | bf16_mfu: 53.80 | total time: 85.39m | eta: 25.4m step 04295/05568 (77.14%) | loss: 2.436468 | lrm: 0.38 | dt: 1196.01ms | tok/sec: 876,727 | bf16_mfu: 53.86 | total time: 85.41m | eta: 25.4m step 04296/05568 (77.16%) | loss: 2.427484 | lrm: 0.38 | dt: 1193.18ms | tok/sec: 878,808 | bf16_mfu: 53.98 | total time: 85.43m | eta: 25.3m step 04297/05568 (77.17%) | loss: 2.433363 | lrm: 0.38 | dt: 1195.70ms | tok/sec: 876,957 | bf16_mfu: 53.87 | total time: 85.45m | eta: 25.3m step 04298/05568 (77.19%) | loss: 2.437564 | lrm: 0.38 | dt: 1195.44ms | tok/sec: 877,149 | bf16_mfu: 53.88 | total time: 85.47m | eta: 25.3m step 04299/05568 (77.21%) | loss: 2.444963 | lrm: 0.38 | dt: 1190.12ms | tok/sec: 881,070 | bf16_mfu: 54.12 | total time: 85.49m | eta: 25.3m step 04300/05568 (77.23%) | loss: 2.435586 | lrm: 0.38 | dt: 1197.17ms | tok/sec: 875,880 | bf16_mfu: 53.80 | total time: 85.51m | eta: 25.3m step 04301/05568 (77.24%) | loss: 2.434406 | lrm: 0.38 | dt: 1193.21ms | tok/sec: 878,787 | bf16_mfu: 53.98 | total time: 85.53m | eta: 25.2m step 04302/05568 (77.26%) | loss: 2.443302 | lrm: 0.38 | dt: 1193.20ms | tok/sec: 878,795 | bf16_mfu: 53.98 | total time: 85.55m | eta: 25.2m step 04303/05568 (77.28%) | loss: 2.432027 | lrm: 0.38 | dt: 1192.66ms | tok/sec: 879,193 | bf16_mfu: 54.01 | total time: 85.57m | eta: 25.2m step 04304/05568 (77.30%) | loss: 2.435865 | lrm: 0.38 | dt: 1198.60ms | tok/sec: 874,834 | bf16_mfu: 53.74 | total time: 85.59m | eta: 25.2m step 04305/05568 (77.32%) | loss: 2.429785 | lrm: 0.38 | dt: 1189.97ms | tok/sec: 881,175 | bf16_mfu: 54.13 | total time: 85.61m | eta: 25.2m step 04306/05568 (77.33%) | loss: 2.434547 | lrm: 0.38 | dt: 1191.09ms | tok/sec: 880,346 | bf16_mfu: 54.08 | total time: 85.63m | eta: 25.1m step 04307/05568 (77.35%) | loss: 2.411240 | lrm: 0.38 | dt: 1197.84ms | tok/sec: 875,391 | bf16_mfu: 53.77 | total time: 85.65m | eta: 25.1m step 04308/05568 (77.37%) | loss: 2.407965 | lrm: 0.38 | dt: 1198.07ms | tok/sec: 875,222 | bf16_mfu: 53.76 | total time: 85.67m | eta: 25.1m step 04309/05568 (77.39%) | loss: 2.427611 | lrm: 0.38 | dt: 1196.27ms | tok/sec: 876,540 | bf16_mfu: 53.84 | total time: 85.69m | eta: 25.1m step 04310/05568 (77.41%) | loss: 2.429613 | lrm: 0.38 | dt: 1195.47ms | tok/sec: 877,124 | bf16_mfu: 53.88 | total time: 85.71m | eta: 25.1m step 04311/05568 (77.42%) | loss: 2.426367 | lrm: 0.38 | dt: 1190.77ms | tok/sec: 880,586 | bf16_mfu: 54.09 | total time: 85.73m | eta: 25.0m step 04312/05568 (77.44%) | loss: 2.415896 | lrm: 0.38 | dt: 1195.82ms | tok/sec: 876,868 | bf16_mfu: 53.86 | total time: 85.75m | eta: 25.0m step 04313/05568 (77.46%) | loss: 2.416341 | lrm: 0.38 | dt: 1194.00ms | tok/sec: 878,205 | bf16_mfu: 53.95 | total time: 85.77m | eta: 25.0m step 04314/05568 (77.48%) | loss: 2.411253 | lrm: 0.38 | dt: 1194.34ms | tok/sec: 877,956 | bf16_mfu: 53.93 | total time: 85.79m | eta: 25.0m step 04315/05568 (77.50%) | loss: 2.420268 | lrm: 0.38 | dt: 1191.97ms | tok/sec: 879,702 | bf16_mfu: 54.04 | total time: 85.81m | eta: 25.0m step 04316/05568 (77.51%) | loss: 2.427963 | lrm: 0.38 | dt: 1201.76ms | tok/sec: 872,530 | bf16_mfu: 53.60 | total time: 85.83m | eta: 24.9m step 04317/05568 (77.53%) | loss: 2.428282 | lrm: 0.38 | dt: 1192.57ms | tok/sec: 879,255 | bf16_mfu: 54.01 | total time: 85.85m | eta: 24.9m step 04318/05568 (77.55%) | loss: 2.426357 | lrm: 0.38 | dt: 1198.10ms | tok/sec: 875,195 | bf16_mfu: 53.76 | total time: 85.87m | eta: 24.9m step 04319/05568 (77.57%) | loss: 2.427541 | lrm: 0.38 | dt: 1195.03ms | tok/sec: 877,450 | bf16_mfu: 53.90 | total time: 85.89m | eta: 24.9m step 04320/05568 (77.59%) | loss: 2.414354 | lrm: 0.38 | dt: 1192.92ms | tok/sec: 878,996 | bf16_mfu: 53.99 | total time: 85.91m | eta: 24.9m step 04321/05568 (77.60%) | loss: 2.413994 | lrm: 0.38 | dt: 1196.48ms | tok/sec: 876,387 | bf16_mfu: 53.83 | total time: 85.93m | eta: 24.8m step 04322/05568 (77.62%) | loss: 2.423778 | lrm: 0.38 | dt: 1193.01ms | tok/sec: 878,936 | bf16_mfu: 53.99 | total time: 85.95m | eta: 24.8m step 04323/05568 (77.64%) | loss: 2.431985 | lrm: 0.38 | dt: 1199.94ms | tok/sec: 873,853 | bf16_mfu: 53.68 | total time: 85.97m | eta: 24.8m step 04324/05568 (77.66%) | loss: 2.434507 | lrm: 0.38 | dt: 1196.19ms | tok/sec: 876,596 | bf16_mfu: 53.85 | total time: 85.99m | eta: 24.8m step 04325/05568 (77.68%) | loss: 2.436235 | lrm: 0.38 | dt: 1198.13ms | tok/sec: 875,174 | bf16_mfu: 53.76 | total time: 86.01m | eta: 24.8m step 04326/05568 (77.69%) | loss: 2.432964 | lrm: 0.38 | dt: 1195.30ms | tok/sec: 877,246 | bf16_mfu: 53.89 | total time: 86.03m | eta: 24.7m step 04327/05568 (77.71%) | loss: 2.439091 | lrm: 0.38 | dt: 1194.92ms | tok/sec: 877,525 | bf16_mfu: 53.90 | total time: 86.05m | eta: 24.7m step 04328/05568 (77.73%) | loss: 2.441196 | lrm: 0.38 | dt: 1193.71ms | tok/sec: 878,417 | bf16_mfu: 53.96 | total time: 86.07m | eta: 24.7m step 04329/05568 (77.75%) | loss: 2.438621 | lrm: 0.38 | dt: 1195.36ms | tok/sec: 877,208 | bf16_mfu: 53.88 | total time: 86.09m | eta: 24.7m step 04330/05568 (77.77%) | loss: 2.455005 | lrm: 0.37 | dt: 1194.52ms | tok/sec: 877,823 | bf16_mfu: 53.92 | total time: 86.11m | eta: 24.7m step 04331/05568 (77.78%) | loss: 2.447566 | lrm: 0.37 | dt: 1193.59ms | tok/sec: 878,504 | bf16_mfu: 53.96 | total time: 86.13m | eta: 24.6m step 04332/05568 (77.80%) | loss: 2.440405 | lrm: 0.37 | dt: 1197.73ms | tok/sec: 875,471 | bf16_mfu: 53.78 | total time: 86.15m | eta: 24.6m step 04333/05568 (77.82%) | loss: 2.441661 | lrm: 0.37 | dt: 1190.63ms | tok/sec: 880,692 | bf16_mfu: 54.10 | total time: 86.17m | eta: 24.6m step 04334/05568 (77.84%) | loss: 2.453458 | lrm: 0.37 | dt: 1191.37ms | tok/sec: 880,145 | bf16_mfu: 54.07 | total time: 86.19m | eta: 24.6m step 04335/05568 (77.86%) | loss: 2.455833 | lrm: 0.37 | dt: 1193.59ms | tok/sec: 878,507 | bf16_mfu: 53.96 | total time: 86.21m | eta: 24.6m step 04336/05568 (77.87%) | loss: 2.446683 | lrm: 0.37 | dt: 1197.14ms | tok/sec: 875,903 | bf16_mfu: 53.80 | total time: 86.23m | eta: 24.5m step 04337/05568 (77.89%) | loss: 2.444698 | lrm: 0.37 | dt: 1193.55ms | tok/sec: 878,533 | bf16_mfu: 53.97 | total time: 86.25m | eta: 24.5m step 04338/05568 (77.91%) | loss: 2.442086 | lrm: 0.37 | dt: 1192.21ms | tok/sec: 879,526 | bf16_mfu: 54.03 | total time: 86.27m | eta: 24.5m step 04339/05568 (77.93%) | loss: 2.441448 | lrm: 0.37 | dt: 1198.74ms | tok/sec: 874,735 | bf16_mfu: 53.73 | total time: 86.29m | eta: 24.5m step 04340/05568 (77.95%) | loss: 2.432521 | lrm: 0.37 | dt: 1195.73ms | tok/sec: 876,936 | bf16_mfu: 53.87 | total time: 86.31m | eta: 24.5m step 04341/05568 (77.96%) | loss: 2.438555 | lrm: 0.37 | dt: 1193.44ms | tok/sec: 878,613 | bf16_mfu: 53.97 | total time: 86.33m | eta: 24.4m step 04342/05568 (77.98%) | loss: 2.430304 | lrm: 0.37 | dt: 1190.06ms | tok/sec: 881,109 | bf16_mfu: 54.12 | total time: 86.35m | eta: 24.4m step 04343/05568 (78.00%) | loss: 2.427787 | lrm: 0.37 | dt: 1196.68ms | tok/sec: 876,236 | bf16_mfu: 53.83 | total time: 86.37m | eta: 24.4m step 04344/05568 (78.02%) | loss: 2.428269 | lrm: 0.37 | dt: 1200.32ms | tok/sec: 873,579 | bf16_mfu: 53.66 | total time: 86.39m | eta: 24.4m step 04345/05568 (78.04%) | loss: 2.438301 | lrm: 0.37 | dt: 1195.17ms | tok/sec: 877,345 | bf16_mfu: 53.89 | total time: 86.41m | eta: 24.4m step 04346/05568 (78.05%) | loss: 2.434030 | lrm: 0.37 | dt: 1198.04ms | tok/sec: 875,243 | bf16_mfu: 53.76 | total time: 86.43m | eta: 24.3m step 04347/05568 (78.07%) | loss: 2.427756 | lrm: 0.37 | dt: 1187.62ms | tok/sec: 882,919 | bf16_mfu: 54.24 | total time: 86.45m | eta: 24.3m step 04348/05568 (78.09%) | loss: 2.424054 | lrm: 0.37 | dt: 1192.87ms | tok/sec: 879,038 | bf16_mfu: 54.00 | total time: 86.47m | eta: 24.3m step 04349/05568 (78.11%) | loss: 2.424594 | lrm: 0.37 | dt: 1194.45ms | tok/sec: 877,876 | bf16_mfu: 53.93 | total time: 86.49m | eta: 24.3m step 04350/05568 (78.12%) | loss: 2.421668 | lrm: 0.37 | dt: 1198.61ms | tok/sec: 874,825 | bf16_mfu: 53.74 | total time: 86.51m | eta: 24.3m step 04351/05568 (78.14%) | loss: 2.424974 | lrm: 0.37 | dt: 1196.06ms | tok/sec: 876,689 | bf16_mfu: 53.85 | total time: 86.53m | eta: 24.2m step 04352/05568 (78.16%) | loss: 2.432972 | lrm: 0.37 | dt: 1195.23ms | tok/sec: 877,298 | bf16_mfu: 53.89 | total time: 86.55m | eta: 24.2m step 04353/05568 (78.18%) | loss: 2.423966 | lrm: 0.37 | dt: 1198.23ms | tok/sec: 875,101 | bf16_mfu: 53.76 | total time: 86.57m | eta: 24.2m step 04354/05568 (78.20%) | loss: 2.436703 | lrm: 0.37 | dt: 1199.01ms | tok/sec: 874,531 | bf16_mfu: 53.72 | total time: 86.59m | eta: 24.2m step 04355/05568 (78.21%) | loss: 2.436041 | lrm: 0.37 | dt: 1194.31ms | tok/sec: 877,979 | bf16_mfu: 53.93 | total time: 86.61m | eta: 24.2m step 04356/05568 (78.23%) | loss: 2.432691 | lrm: 0.37 | dt: 1192.30ms | tok/sec: 879,455 | bf16_mfu: 54.02 | total time: 86.63m | eta: 24.1m step 04357/05568 (78.25%) | loss: 2.425724 | lrm: 0.37 | dt: 1196.51ms | tok/sec: 876,362 | bf16_mfu: 53.83 | total time: 86.65m | eta: 24.1m step 04358/05568 (78.27%) | loss: 2.421944 | lrm: 0.37 | dt: 1193.35ms | tok/sec: 878,680 | bf16_mfu: 53.98 | total time: 86.67m | eta: 24.1m step 04359/05568 (78.29%) | loss: 2.433102 | lrm: 0.37 | dt: 1192.95ms | tok/sec: 878,977 | bf16_mfu: 53.99 | total time: 86.69m | eta: 24.1m step 04360/05568 (78.30%) | loss: 2.415914 | lrm: 0.37 | dt: 1193.24ms | tok/sec: 878,765 | bf16_mfu: 53.98 | total time: 86.71m | eta: 24.1m step 04361/05568 (78.32%) | loss: 2.417007 | lrm: 0.37 | dt: 1195.47ms | tok/sec: 877,123 | bf16_mfu: 53.88 | total time: 86.73m | eta: 24.0m step 04362/05568 (78.34%) | loss: 2.416407 | lrm: 0.37 | dt: 1193.51ms | tok/sec: 878,564 | bf16_mfu: 53.97 | total time: 86.75m | eta: 24.0m step 04363/05568 (78.36%) | loss: 2.418073 | lrm: 0.37 | dt: 1196.55ms | tok/sec: 876,334 | bf16_mfu: 53.83 | total time: 86.77m | eta: 24.0m step 04364/05568 (78.38%) | loss: 2.427024 | lrm: 0.37 | dt: 1193.14ms | tok/sec: 878,839 | bf16_mfu: 53.99 | total time: 86.78m | eta: 24.0m step 04365/05568 (78.39%) | loss: 2.419163 | lrm: 0.37 | dt: 1199.16ms | tok/sec: 874,426 | bf16_mfu: 53.71 | total time: 86.80m | eta: 24.0m step 04366/05568 (78.41%) | loss: 2.420098 | lrm: 0.37 | dt: 1193.33ms | tok/sec: 878,694 | bf16_mfu: 53.98 | total time: 86.82m | eta: 23.9m step 04367/05568 (78.43%) | loss: 2.421167 | lrm: 0.37 | dt: 1197.56ms | tok/sec: 875,590 | bf16_mfu: 53.79 | total time: 86.84m | eta: 23.9m step 04368/05568 (78.45%) | loss: 2.421653 | lrm: 0.37 | dt: 1194.98ms | tok/sec: 877,482 | bf16_mfu: 53.90 | total time: 86.86m | eta: 23.9m step 04369/05568 (78.47%) | loss: 2.422251 | lrm: 0.36 | dt: 1195.44ms | tok/sec: 877,145 | bf16_mfu: 53.88 | total time: 86.88m | eta: 23.9m step 04370/05568 (78.48%) | loss: 2.421768 | lrm: 0.36 | dt: 1195.13ms | tok/sec: 877,376 | bf16_mfu: 53.90 | total time: 86.90m | eta: 23.9m step 04371/05568 (78.50%) | loss: 2.432596 | lrm: 0.36 | dt: 1191.82ms | tok/sec: 879,814 | bf16_mfu: 54.04 | total time: 86.92m | eta: 23.8m step 04372/05568 (78.52%) | loss: 2.435920 | lrm: 0.36 | dt: 1194.15ms | tok/sec: 878,091 | bf16_mfu: 53.94 | total time: 86.94m | eta: 23.8m step 04373/05568 (78.54%) | loss: 2.442121 | lrm: 0.36 | dt: 1195.63ms | tok/sec: 877,007 | bf16_mfu: 53.87 | total time: 86.96m | eta: 23.8m step 04374/05568 (78.56%) | loss: 2.432593 | lrm: 0.36 | dt: 1196.29ms | tok/sec: 876,521 | bf16_mfu: 53.84 | total time: 86.98m | eta: 23.8m step 04375/05568 (78.57%) | loss: 2.438080 | lrm: 0.36 | dt: 1194.74ms | tok/sec: 877,663 | bf16_mfu: 53.91 | total time: 87.00m | eta: 23.8m step 04376/05568 (78.59%) | loss: 2.450869 | lrm: 0.36 | dt: 1198.73ms | tok/sec: 874,739 | bf16_mfu: 53.73 | total time: 87.02m | eta: 23.7m step 04377/05568 (78.61%) | loss: 2.444614 | lrm: 0.36 | dt: 1194.72ms | tok/sec: 877,672 | bf16_mfu: 53.91 | total time: 87.04m | eta: 23.7m step 04378/05568 (78.63%) | loss: 2.449855 | lrm: 0.36 | dt: 1198.10ms | tok/sec: 875,196 | bf16_mfu: 53.76 | total time: 87.06m | eta: 23.7m step 04379/05568 (78.65%) | loss: 2.451045 | lrm: 0.36 | dt: 1199.63ms | tok/sec: 874,082 | bf16_mfu: 53.69 | total time: 87.08m | eta: 23.7m step 04380/05568 (78.66%) | loss: 2.446490 | lrm: 0.36 | dt: 1196.62ms | tok/sec: 876,279 | bf16_mfu: 53.83 | total time: 87.10m | eta: 23.7m step 04381/05568 (78.68%) | loss: 2.452722 | lrm: 0.36 | dt: 1197.33ms | tok/sec: 875,758 | bf16_mfu: 53.80 | total time: 87.12m | eta: 23.6m step 04382/05568 (78.70%) | loss: 2.444882 | lrm: 0.36 | dt: 1196.31ms | tok/sec: 876,508 | bf16_mfu: 53.84 | total time: 87.14m | eta: 23.6m step 04383/05568 (78.72%) | loss: 2.441585 | lrm: 0.36 | dt: 1190.97ms | tok/sec: 880,435 | bf16_mfu: 54.08 | total time: 87.16m | eta: 23.6m step 04384/05568 (78.74%) | loss: 2.436783 | lrm: 0.36 | dt: 1195.04ms | tok/sec: 877,437 | bf16_mfu: 53.90 | total time: 87.18m | eta: 23.6m step 04385/05568 (78.75%) | loss: 2.413800 | lrm: 0.36 | dt: 1195.56ms | tok/sec: 877,058 | bf16_mfu: 53.88 | total time: 87.20m | eta: 23.6m step 04386/05568 (78.77%) | loss: 2.420057 | lrm: 0.36 | dt: 1202.64ms | tok/sec: 871,893 | bf16_mfu: 53.56 | total time: 87.22m | eta: 23.5m step 04387/05568 (78.79%) | loss: 2.423634 | lrm: 0.36 | dt: 1195.44ms | tok/sec: 877,147 | bf16_mfu: 53.88 | total time: 87.24m | eta: 23.5m step 04388/05568 (78.81%) | loss: 2.420618 | lrm: 0.36 | dt: 1191.02ms | tok/sec: 880,397 | bf16_mfu: 54.08 | total time: 87.26m | eta: 23.5m step 04389/05568 (78.83%) | loss: 2.412132 | lrm: 0.36 | dt: 1193.38ms | tok/sec: 878,657 | bf16_mfu: 53.97 | total time: 87.28m | eta: 23.5m step 04390/05568 (78.84%) | loss: 2.417335 | lrm: 0.36 | dt: 1197.87ms | tok/sec: 875,367 | bf16_mfu: 53.77 | total time: 87.30m | eta: 23.5m step 04391/05568 (78.86%) | loss: 2.419415 | lrm: 0.36 | dt: 1189.87ms | tok/sec: 881,253 | bf16_mfu: 54.13 | total time: 87.32m | eta: 23.4m step 04392/05568 (78.88%) | loss: 2.429277 | lrm: 0.36 | dt: 1196.52ms | tok/sec: 876,354 | bf16_mfu: 53.83 | total time: 87.34m | eta: 23.4m step 04393/05568 (78.90%) | loss: 2.433339 | lrm: 0.36 | dt: 1194.66ms | tok/sec: 877,718 | bf16_mfu: 53.92 | total time: 87.36m | eta: 23.4m step 04394/05568 (78.92%) | loss: 2.435870 | lrm: 0.36 | dt: 1194.50ms | tok/sec: 877,836 | bf16_mfu: 53.92 | total time: 87.38m | eta: 23.4m step 04395/05568 (78.93%) | loss: 2.446465 | lrm: 0.36 | dt: 1193.24ms | tok/sec: 878,764 | bf16_mfu: 53.98 | total time: 87.40m | eta: 23.4m step 04396/05568 (78.95%) | loss: 2.442270 | lrm: 0.36 | dt: 1195.43ms | tok/sec: 877,153 | bf16_mfu: 53.88 | total time: 87.42m | eta: 23.3m step 04397/05568 (78.97%) | loss: 2.445429 | lrm: 0.36 | dt: 1193.63ms | tok/sec: 878,473 | bf16_mfu: 53.96 | total time: 87.44m | eta: 23.3m step 04398/05568 (78.99%) | loss: 2.445232 | lrm: 0.36 | dt: 1197.50ms | tok/sec: 875,636 | bf16_mfu: 53.79 | total time: 87.46m | eta: 23.3m step 04399/05568 (79.01%) | loss: 2.448204 | lrm: 0.36 | dt: 1192.85ms | tok/sec: 879,053 | bf16_mfu: 54.00 | total time: 87.48m | eta: 23.3m step 04400/05568 (79.02%) | loss: 2.447897 | lrm: 0.36 | dt: 1194.57ms | tok/sec: 877,783 | bf16_mfu: 53.92 | total time: 87.50m | eta: 23.3m step 04401/05568 (79.04%) | loss: 2.449540 | lrm: 0.36 | dt: 1194.62ms | tok/sec: 877,749 | bf16_mfu: 53.92 | total time: 87.52m | eta: 23.2m step 04402/05568 (79.06%) | loss: 2.449249 | lrm: 0.36 | dt: 1196.27ms | tok/sec: 876,537 | bf16_mfu: 53.84 | total time: 87.54m | eta: 23.2m step 04403/05568 (79.08%) | loss: 2.446340 | lrm: 0.36 | dt: 1193.74ms | tok/sec: 878,392 | bf16_mfu: 53.96 | total time: 87.56m | eta: 23.2m step 04404/05568 (79.09%) | loss: 2.442247 | lrm: 0.36 | dt: 1196.71ms | tok/sec: 876,218 | bf16_mfu: 53.82 | total time: 87.58m | eta: 23.2m step 04405/05568 (79.11%) | loss: 2.452769 | lrm: 0.36 | dt: 1196.92ms | tok/sec: 876,059 | bf16_mfu: 53.81 | total time: 87.60m | eta: 23.2m step 04406/05568 (79.13%) | loss: 2.454296 | lrm: 0.36 | dt: 1196.32ms | tok/sec: 876,504 | bf16_mfu: 53.84 | total time: 87.62m | eta: 23.1m step 04407/05568 (79.15%) | loss: 2.448461 | lrm: 0.35 | dt: 1196.26ms | tok/sec: 876,542 | bf16_mfu: 53.84 | total time: 87.64m | eta: 23.1m step 04408/05568 (79.17%) | loss: 2.451126 | lrm: 0.35 | dt: 1195.33ms | tok/sec: 877,226 | bf16_mfu: 53.89 | total time: 87.66m | eta: 23.1m step 04409/05568 (79.18%) | loss: 2.440176 | lrm: 0.35 | dt: 1193.97ms | tok/sec: 878,223 | bf16_mfu: 53.95 | total time: 87.68m | eta: 23.1m step 04410/05568 (79.20%) | loss: 2.433175 | lrm: 0.35 | dt: 1197.98ms | tok/sec: 875,283 | bf16_mfu: 53.77 | total time: 87.70m | eta: 23.1m step 04411/05568 (79.22%) | loss: 2.433155 | lrm: 0.35 | dt: 1195.45ms | tok/sec: 877,137 | bf16_mfu: 53.88 | total time: 87.72m | eta: 23.0m step 04412/05568 (79.24%) | loss: 2.427673 | lrm: 0.35 | dt: 1197.62ms | tok/sec: 875,550 | bf16_mfu: 53.78 | total time: 87.74m | eta: 23.0m step 04413/05568 (79.26%) | loss: 2.434693 | lrm: 0.35 | dt: 1194.13ms | tok/sec: 878,111 | bf16_mfu: 53.94 | total time: 87.76m | eta: 23.0m step 04414/05568 (79.27%) | loss: 2.455208 | lrm: 0.35 | dt: 1190.90ms | tok/sec: 880,491 | bf16_mfu: 54.09 | total time: 87.78m | eta: 23.0m step 04415/05568 (79.29%) | loss: 2.457476 | lrm: 0.35 | dt: 1192.55ms | tok/sec: 879,273 | bf16_mfu: 54.01 | total time: 87.80m | eta: 23.0m step 04416/05568 (79.31%) | loss: 2.454113 | lrm: 0.35 | dt: 1191.48ms | tok/sec: 880,064 | bf16_mfu: 54.06 | total time: 87.82m | eta: 22.9m step 04417/05568 (79.33%) | loss: 2.446807 | lrm: 0.35 | dt: 1195.04ms | tok/sec: 877,441 | bf16_mfu: 53.90 | total time: 87.84m | eta: 22.9m step 04418/05568 (79.35%) | loss: 2.455492 | lrm: 0.35 | dt: 1197.04ms | tok/sec: 875,975 | bf16_mfu: 53.81 | total time: 87.86m | eta: 22.9m step 04419/05568 (79.36%) | loss: 2.455071 | lrm: 0.35 | dt: 1196.37ms | tok/sec: 876,466 | bf16_mfu: 53.84 | total time: 87.88m | eta: 22.9m step 04420/05568 (79.38%) | loss: 2.458931 | lrm: 0.35 | dt: 1191.26ms | tok/sec: 880,226 | bf16_mfu: 54.07 | total time: 87.90m | eta: 22.9m step 04421/05568 (79.40%) | loss: 2.457614 | lrm: 0.35 | dt: 1194.21ms | tok/sec: 878,051 | bf16_mfu: 53.94 | total time: 87.92m | eta: 22.8m step 04422/05568 (79.42%) | loss: 2.457376 | lrm: 0.35 | dt: 1198.42ms | tok/sec: 874,963 | bf16_mfu: 53.75 | total time: 87.94m | eta: 22.8m step 04423/05568 (79.44%) | loss: 2.441486 | lrm: 0.35 | dt: 1195.13ms | tok/sec: 877,374 | bf16_mfu: 53.90 | total time: 87.96m | eta: 22.8m step 04424/05568 (79.45%) | loss: 2.427718 | lrm: 0.35 | dt: 1192.24ms | tok/sec: 879,498 | bf16_mfu: 54.03 | total time: 87.98m | eta: 22.8m step 04425/05568 (79.47%) | loss: 2.435268 | lrm: 0.35 | dt: 1195.70ms | tok/sec: 876,959 | bf16_mfu: 53.87 | total time: 88.00m | eta: 22.8m step 04426/05568 (79.49%) | loss: 2.429645 | lrm: 0.35 | dt: 1191.09ms | tok/sec: 880,349 | bf16_mfu: 54.08 | total time: 88.02m | eta: 22.7m step 04427/05568 (79.51%) | loss: 2.430982 | lrm: 0.35 | dt: 1197.60ms | tok/sec: 875,563 | bf16_mfu: 53.78 | total time: 88.04m | eta: 22.7m step 04428/05568 (79.53%) | loss: 2.426722 | lrm: 0.35 | dt: 1193.14ms | tok/sec: 878,836 | bf16_mfu: 53.98 | total time: 88.06m | eta: 22.7m step 04429/05568 (79.54%) | loss: 2.422460 | lrm: 0.35 | dt: 1198.82ms | tok/sec: 874,672 | bf16_mfu: 53.73 | total time: 88.08m | eta: 22.7m step 04430/05568 (79.56%) | loss: 2.419532 | lrm: 0.35 | dt: 1197.73ms | tok/sec: 875,466 | bf16_mfu: 53.78 | total time: 88.10m | eta: 22.7m step 04431/05568 (79.58%) | loss: 2.416727 | lrm: 0.35 | dt: 1194.87ms | tok/sec: 877,568 | bf16_mfu: 53.91 | total time: 88.12m | eta: 22.6m step 04432/05568 (79.60%) | loss: 2.414378 | lrm: 0.35 | dt: 1198.28ms | tok/sec: 875,065 | bf16_mfu: 53.75 | total time: 88.14m | eta: 22.6m step 04433/05568 (79.62%) | loss: 2.419864 | lrm: 0.35 | dt: 1196.39ms | tok/sec: 876,447 | bf16_mfu: 53.84 | total time: 88.16m | eta: 22.6m step 04434/05568 (79.63%) | loss: 2.420880 | lrm: 0.35 | dt: 1192.93ms | tok/sec: 878,992 | bf16_mfu: 53.99 | total time: 88.18m | eta: 22.6m step 04435/05568 (79.65%) | loss: 2.414829 | lrm: 0.35 | dt: 1195.94ms | tok/sec: 876,779 | bf16_mfu: 53.86 | total time: 88.20m | eta: 22.6m step 04436/05568 (79.67%) | loss: 2.413930 | lrm: 0.35 | dt: 1194.88ms | tok/sec: 877,555 | bf16_mfu: 53.91 | total time: 88.22m | eta: 22.5m step 04437/05568 (79.69%) | loss: 2.424176 | lrm: 0.35 | dt: 1195.30ms | tok/sec: 877,247 | bf16_mfu: 53.89 | total time: 88.24m | eta: 22.5m step 04438/05568 (79.71%) | loss: 2.419629 | lrm: 0.35 | dt: 1192.76ms | tok/sec: 879,119 | bf16_mfu: 54.00 | total time: 88.26m | eta: 22.5m step 04439/05568 (79.72%) | loss: 2.406011 | lrm: 0.35 | dt: 1199.18ms | tok/sec: 874,408 | bf16_mfu: 53.71 | total time: 88.28m | eta: 22.5m step 04440/05568 (79.74%) | loss: 2.414270 | lrm: 0.35 | dt: 1191.87ms | tok/sec: 879,774 | bf16_mfu: 54.04 | total time: 88.30m | eta: 22.5m step 04441/05568 (79.76%) | loss: 2.416438 | lrm: 0.35 | dt: 1195.33ms | tok/sec: 877,226 | bf16_mfu: 53.89 | total time: 88.32m | eta: 22.4m step 04442/05568 (79.78%) | loss: 2.424629 | lrm: 0.35 | dt: 1194.15ms | tok/sec: 878,092 | bf16_mfu: 53.94 | total time: 88.34m | eta: 22.4m step 04443/05568 (79.80%) | loss: 2.427263 | lrm: 0.35 | dt: 1195.49ms | tok/sec: 877,106 | bf16_mfu: 53.88 | total time: 88.36m | eta: 22.4m step 04444/05568 (79.81%) | loss: 2.423200 | lrm: 0.35 | dt: 1196.56ms | tok/sec: 876,323 | bf16_mfu: 53.83 | total time: 88.38m | eta: 22.4m step 04445/05568 (79.83%) | loss: 2.423631 | lrm: 0.34 | dt: 1196.28ms | tok/sec: 876,533 | bf16_mfu: 53.84 | total time: 88.40m | eta: 22.4m step 04446/05568 (79.85%) | loss: 2.430224 | lrm: 0.34 | dt: 1190.16ms | tok/sec: 881,040 | bf16_mfu: 54.12 | total time: 88.42m | eta: 22.3m step 04447/05568 (79.87%) | loss: 2.434865 | lrm: 0.34 | dt: 1191.88ms | tok/sec: 879,762 | bf16_mfu: 54.04 | total time: 88.44m | eta: 22.3m step 04448/05568 (79.89%) | loss: 2.436695 | lrm: 0.34 | dt: 1196.39ms | tok/sec: 876,452 | bf16_mfu: 53.84 | total time: 88.46m | eta: 22.3m step 04449/05568 (79.90%) | loss: 2.440467 | lrm: 0.34 | dt: 1198.23ms | tok/sec: 875,102 | bf16_mfu: 53.76 | total time: 88.48m | eta: 22.3m step 04450/05568 (79.92%) | loss: 2.440856 | lrm: 0.34 | dt: 1192.54ms | tok/sec: 879,276 | bf16_mfu: 54.01 | total time: 88.50m | eta: 22.3m step 04451/05568 (79.94%) | loss: 2.433282 | lrm: 0.34 | dt: 1194.17ms | tok/sec: 878,076 | bf16_mfu: 53.94 | total time: 88.52m | eta: 22.2m step 04452/05568 (79.96%) | loss: 2.433741 | lrm: 0.34 | dt: 1198.22ms | tok/sec: 875,110 | bf16_mfu: 53.76 | total time: 88.54m | eta: 22.2m step 04453/05568 (79.97%) | loss: 2.453231 | lrm: 0.34 | dt: 1193.91ms | tok/sec: 878,269 | bf16_mfu: 53.95 | total time: 88.56m | eta: 22.2m step 04454/05568 (79.99%) | loss: 2.451423 | lrm: 0.34 | dt: 1192.51ms | tok/sec: 879,301 | bf16_mfu: 54.01 | total time: 88.58m | eta: 22.2m step 04455/05568 (80.01%) | loss: 2.455712 | lrm: 0.34 | dt: 1195.46ms | tok/sec: 877,131 | bf16_mfu: 53.88 | total time: 88.60m | eta: 22.2m step 04456/05568 (80.03%) | loss: 2.446683 | lrm: 0.34 | dt: 1190.22ms | tok/sec: 880,993 | bf16_mfu: 54.12 | total time: 88.62m | eta: 22.1m step 04457/05568 (80.05%) | loss: 2.436308 | lrm: 0.34 | dt: 1194.63ms | tok/sec: 877,739 | bf16_mfu: 53.92 | total time: 88.64m | eta: 22.1m step 04458/05568 (80.06%) | loss: 2.430263 | lrm: 0.34 | dt: 1194.26ms | tok/sec: 878,013 | bf16_mfu: 53.93 | total time: 88.66m | eta: 22.1m step 04459/05568 (80.08%) | loss: 2.429005 | lrm: 0.34 | dt: 1192.69ms | tok/sec: 879,165 | bf16_mfu: 54.01 | total time: 88.68m | eta: 22.1m step 04460/05568 (80.10%) | loss: 2.425194 | lrm: 0.34 | dt: 1192.21ms | tok/sec: 879,522 | bf16_mfu: 54.03 | total time: 88.70m | eta: 22.1m step 04461/05568 (80.12%) | loss: 2.425118 | lrm: 0.34 | dt: 1196.30ms | tok/sec: 876,513 | bf16_mfu: 53.84 | total time: 88.72m | eta: 22.0m step 04462/05568 (80.14%) | loss: 2.433522 | lrm: 0.34 | dt: 1193.54ms | tok/sec: 878,545 | bf16_mfu: 53.97 | total time: 88.74m | eta: 22.0m step 04463/05568 (80.15%) | loss: 2.427878 | lrm: 0.34 | dt: 1191.46ms | tok/sec: 880,076 | bf16_mfu: 54.06 | total time: 88.76m | eta: 22.0m step 04464/05568 (80.17%) | loss: 2.431233 | lrm: 0.34 | dt: 1197.78ms | tok/sec: 875,430 | bf16_mfu: 53.78 | total time: 88.78m | eta: 22.0m step 04465/05568 (80.19%) | loss: 2.433221 | lrm: 0.34 | dt: 1192.83ms | tok/sec: 879,066 | bf16_mfu: 54.00 | total time: 88.80m | eta: 22.0m step 04466/05568 (80.21%) | loss: 2.430036 | lrm: 0.34 | dt: 1195.40ms | tok/sec: 877,176 | bf16_mfu: 53.88 | total time: 88.82m | eta: 21.9m step 04467/05568 (80.23%) | loss: 2.425179 | lrm: 0.34 | dt: 1191.06ms | tok/sec: 880,375 | bf16_mfu: 54.08 | total time: 88.84m | eta: 21.9m step 04468/05568 (80.24%) | loss: 2.417440 | lrm: 0.34 | dt: 1197.17ms | tok/sec: 875,876 | bf16_mfu: 53.80 | total time: 88.86m | eta: 21.9m step 04469/05568 (80.26%) | loss: 2.415755 | lrm: 0.34 | dt: 1201.87ms | tok/sec: 872,452 | bf16_mfu: 53.59 | total time: 88.88m | eta: 21.9m step 04470/05568 (80.28%) | loss: 2.424637 | lrm: 0.34 | dt: 1200.97ms | tok/sec: 873,110 | bf16_mfu: 53.63 | total time: 88.90m | eta: 21.9m step 04471/05568 (80.30%) | loss: 2.428936 | lrm: 0.34 | dt: 1191.42ms | tok/sec: 880,107 | bf16_mfu: 54.06 | total time: 88.92m | eta: 21.8m step 04472/05568 (80.32%) | loss: 2.433071 | lrm: 0.34 | dt: 1193.18ms | tok/sec: 878,808 | bf16_mfu: 53.98 | total time: 88.94m | eta: 21.8m step 04473/05568 (80.33%) | loss: 2.430090 | lrm: 0.34 | dt: 1194.46ms | tok/sec: 877,865 | bf16_mfu: 53.93 | total time: 88.96m | eta: 21.8m step 04474/05568 (80.35%) | loss: 2.423744 | lrm: 0.34 | dt: 1193.89ms | tok/sec: 878,287 | bf16_mfu: 53.95 | total time: 88.98m | eta: 21.8m step 04475/05568 (80.37%) | loss: 2.436055 | lrm: 0.34 | dt: 1190.53ms | tok/sec: 880,762 | bf16_mfu: 54.10 | total time: 89.00m | eta: 21.8m step 04476/05568 (80.39%) | loss: 2.434357 | lrm: 0.34 | dt: 1202.06ms | tok/sec: 872,315 | bf16_mfu: 53.58 | total time: 89.02m | eta: 21.7m step 04477/05568 (80.41%) | loss: 2.441298 | lrm: 0.34 | dt: 1194.73ms | tok/sec: 877,664 | bf16_mfu: 53.91 | total time: 89.04m | eta: 21.7m step 04478/05568 (80.42%) | loss: 2.435595 | lrm: 0.34 | dt: 1197.79ms | tok/sec: 875,427 | bf16_mfu: 53.78 | total time: 89.06m | eta: 21.7m step 04479/05568 (80.44%) | loss: 2.432271 | lrm: 0.34 | dt: 1194.06ms | tok/sec: 878,161 | bf16_mfu: 53.94 | total time: 89.08m | eta: 21.7m step 04480/05568 (80.46%) | loss: 2.427373 | lrm: 0.34 | dt: 1192.59ms | tok/sec: 879,239 | bf16_mfu: 54.01 | total time: 89.10m | eta: 21.7m step 04481/05568 (80.48%) | loss: 2.416463 | lrm: 0.34 | dt: 1198.08ms | tok/sec: 875,215 | bf16_mfu: 53.76 | total time: 89.12m | eta: 21.6m step 04482/05568 (80.50%) | loss: 2.400550 | lrm: 0.34 | dt: 1195.25ms | tok/sec: 877,287 | bf16_mfu: 53.89 | total time: 89.14m | eta: 21.6m step 04483/05568 (80.51%) | loss: 2.410794 | lrm: 0.33 | dt: 1192.64ms | tok/sec: 879,203 | bf16_mfu: 54.01 | total time: 89.16m | eta: 21.6m step 04484/05568 (80.53%) | loss: 2.407821 | lrm: 0.33 | dt: 1195.09ms | tok/sec: 877,406 | bf16_mfu: 53.90 | total time: 89.18m | eta: 21.6m step 04485/05568 (80.55%) | loss: 2.408456 | lrm: 0.33 | dt: 1194.34ms | tok/sec: 877,954 | bf16_mfu: 53.93 | total time: 89.20m | eta: 21.6m step 04486/05568 (80.57%) | loss: 2.407967 | lrm: 0.33 | dt: 1194.71ms | tok/sec: 877,684 | bf16_mfu: 53.91 | total time: 89.21m | eta: 21.5m step 04487/05568 (80.59%) | loss: 2.406644 | lrm: 0.33 | dt: 1197.12ms | tok/sec: 875,915 | bf16_mfu: 53.81 | total time: 89.23m | eta: 21.5m step 04488/05568 (80.60%) | loss: 2.403949 | lrm: 0.33 | dt: 1194.45ms | tok/sec: 877,874 | bf16_mfu: 53.93 | total time: 89.25m | eta: 21.5m step 04489/05568 (80.62%) | loss: 2.398312 | lrm: 0.33 | dt: 1194.37ms | tok/sec: 877,928 | bf16_mfu: 53.93 | total time: 89.27m | eta: 21.5m step 04490/05568 (80.64%) | loss: 2.395391 | lrm: 0.33 | dt: 1196.65ms | tok/sec: 876,259 | bf16_mfu: 53.83 | total time: 89.29m | eta: 21.5m step 04491/05568 (80.66%) | loss: 2.400687 | lrm: 0.33 | dt: 1193.23ms | tok/sec: 878,772 | bf16_mfu: 53.98 | total time: 89.31m | eta: 21.4m step 04492/05568 (80.68%) | loss: 2.399263 | lrm: 0.33 | dt: 1198.82ms | tok/sec: 874,675 | bf16_mfu: 53.73 | total time: 89.33m | eta: 21.4m step 04493/05568 (80.69%) | loss: 2.400708 | lrm: 0.33 | dt: 1193.78ms | tok/sec: 878,366 | bf16_mfu: 53.96 | total time: 89.35m | eta: 21.4m step 04494/05568 (80.71%) | loss: 2.406403 | lrm: 0.33 | dt: 1199.55ms | tok/sec: 874,139 | bf16_mfu: 53.70 | total time: 89.37m | eta: 21.4m step 04495/05568 (80.73%) | loss: 2.403672 | lrm: 0.33 | dt: 1195.71ms | tok/sec: 876,946 | bf16_mfu: 53.87 | total time: 89.39m | eta: 21.4m step 04496/05568 (80.75%) | loss: 2.409784 | lrm: 0.33 | dt: 1196.05ms | tok/sec: 876,701 | bf16_mfu: 53.85 | total time: 89.41m | eta: 21.3m step 04497/05568 (80.77%) | loss: 2.413754 | lrm: 0.33 | dt: 1197.15ms | tok/sec: 875,891 | bf16_mfu: 53.80 | total time: 89.43m | eta: 21.3m step 04498/05568 (80.78%) | loss: 2.415260 | lrm: 0.33 | dt: 1193.89ms | tok/sec: 878,287 | bf16_mfu: 53.95 | total time: 89.45m | eta: 21.3m step 04499/05568 (80.80%) | loss: 2.413866 | lrm: 0.33 | dt: 1191.59ms | tok/sec: 879,978 | bf16_mfu: 54.06 | total time: 89.47m | eta: 21.3m step:4500/5568 val_bpb:0.736278 val_time:3.89s step 04500/05568 (80.82%) | loss: 2.399170 | lrm: 0.33 | dt: 1201.88ms | tok/sec: 872,446 | bf16_mfu: 53.59 | total time: 89.49m | eta: 21.3m step 04501/05568 (80.84%) | loss: 2.408624 | lrm: 0.33 | dt: 1193.05ms | tok/sec: 878,907 | bf16_mfu: 53.99 | total time: 89.51m | eta: 21.2m step 04502/05568 (80.85%) | loss: 2.420647 | lrm: 0.33 | dt: 1193.56ms | tok/sec: 878,528 | bf16_mfu: 53.97 | total time: 89.53m | eta: 21.2m step 04503/05568 (80.87%) | loss: 2.416734 | lrm: 0.33 | dt: 1193.71ms | tok/sec: 878,418 | bf16_mfu: 53.96 | total time: 89.55m | eta: 21.2m step 04504/05568 (80.89%) | loss: 2.414212 | lrm: 0.33 | dt: 1195.42ms | tok/sec: 877,157 | bf16_mfu: 53.88 | total time: 89.57m | eta: 21.2m step 04505/05568 (80.91%) | loss: 2.410139 | lrm: 0.33 | dt: 1195.34ms | tok/sec: 877,219 | bf16_mfu: 53.89 | total time: 89.59m | eta: 21.2m step 04506/05568 (80.93%) | loss: 2.409281 | lrm: 0.33 | dt: 1199.09ms | tok/sec: 874,476 | bf16_mfu: 53.72 | total time: 89.61m | eta: 21.1m step 04507/05568 (80.94%) | loss: 2.411930 | lrm: 0.33 | dt: 1188.47ms | tok/sec: 882,290 | bf16_mfu: 54.20 | total time: 89.63m | eta: 21.1m step 04508/05568 (80.96%) | loss: 2.417090 | lrm: 0.33 | dt: 1197.17ms | tok/sec: 875,881 | bf16_mfu: 53.80 | total time: 89.65m | eta: 21.1m step 04509/05568 (80.98%) | loss: 2.420364 | lrm: 0.33 | dt: 1193.50ms | tok/sec: 878,571 | bf16_mfu: 53.97 | total time: 89.67m | eta: 21.1m step 04510/05568 (81.00%) | loss: 2.411513 | lrm: 0.33 | dt: 1199.30ms | tok/sec: 874,325 | bf16_mfu: 53.71 | total time: 89.69m | eta: 21.1m step 04511/05568 (81.02%) | loss: 2.406545 | lrm: 0.33 | dt: 1195.72ms | tok/sec: 876,937 | bf16_mfu: 53.87 | total time: 89.71m | eta: 21.0m step 04512/05568 (81.03%) | loss: 2.399224 | lrm: 0.33 | dt: 1193.37ms | tok/sec: 878,667 | bf16_mfu: 53.97 | total time: 89.73m | eta: 21.0m step 04513/05568 (81.05%) | loss: 2.399576 | lrm: 0.33 | dt: 1195.24ms | tok/sec: 877,293 | bf16_mfu: 53.89 | total time: 89.75m | eta: 21.0m step 04514/05568 (81.07%) | loss: 2.406563 | lrm: 0.33 | dt: 1195.75ms | tok/sec: 876,920 | bf16_mfu: 53.87 | total time: 89.77m | eta: 21.0m step 04515/05568 (81.09%) | loss: 2.409830 | lrm: 0.33 | dt: 1193.52ms | tok/sec: 878,560 | bf16_mfu: 53.97 | total time: 89.79m | eta: 21.0m step 04516/05568 (81.11%) | loss: 2.402816 | lrm: 0.33 | dt: 1196.65ms | tok/sec: 876,256 | bf16_mfu: 53.83 | total time: 89.81m | eta: 20.9m step 04517/05568 (81.12%) | loss: 2.403118 | lrm: 0.33 | dt: 1191.95ms | tok/sec: 879,716 | bf16_mfu: 54.04 | total time: 89.83m | eta: 20.9m step 04518/05568 (81.14%) | loss: 2.400935 | lrm: 0.33 | dt: 1191.13ms | tok/sec: 880,318 | bf16_mfu: 54.08 | total time: 89.85m | eta: 20.9m step 04519/05568 (81.16%) | loss: 2.395346 | lrm: 0.33 | dt: 1198.48ms | tok/sec: 874,924 | bf16_mfu: 53.74 | total time: 89.87m | eta: 20.9m step 04520/05568 (81.18%) | loss: 2.383473 | lrm: 0.33 | dt: 1195.56ms | tok/sec: 877,058 | bf16_mfu: 53.88 | total time: 89.89m | eta: 20.9m step 04521/05568 (81.20%) | loss: 2.395409 | lrm: 0.32 | dt: 1194.04ms | tok/sec: 878,171 | bf16_mfu: 53.94 | total time: 89.91m | eta: 20.8m step 04522/05568 (81.21%) | loss: 2.386871 | lrm: 0.32 | dt: 1193.94ms | tok/sec: 878,245 | bf16_mfu: 53.95 | total time: 89.93m | eta: 20.8m step 04523/05568 (81.23%) | loss: 2.383509 | lrm: 0.32 | dt: 1196.01ms | tok/sec: 876,726 | bf16_mfu: 53.86 | total time: 89.95m | eta: 20.8m step 04524/05568 (81.25%) | loss: 2.373396 | lrm: 0.32 | dt: 1194.01ms | tok/sec: 878,200 | bf16_mfu: 53.95 | total time: 89.97m | eta: 20.8m step 04525/05568 (81.27%) | loss: 2.373681 | lrm: 0.32 | dt: 1191.61ms | tok/sec: 879,968 | bf16_mfu: 54.05 | total time: 89.99m | eta: 20.8m step 04526/05568 (81.29%) | loss: 2.373965 | lrm: 0.32 | dt: 1192.92ms | tok/sec: 879,003 | bf16_mfu: 54.00 | total time: 90.01m | eta: 20.7m step 04527/05568 (81.30%) | loss: 2.378090 | lrm: 0.32 | dt: 1192.84ms | tok/sec: 879,061 | bf16_mfu: 54.00 | total time: 90.03m | eta: 20.7m step 04528/05568 (81.32%) | loss: 2.384427 | lrm: 0.32 | dt: 1197.91ms | tok/sec: 875,341 | bf16_mfu: 53.77 | total time: 90.05m | eta: 20.7m step 04529/05568 (81.34%) | loss: 2.388051 | lrm: 0.32 | dt: 1199.42ms | tok/sec: 874,233 | bf16_mfu: 53.70 | total time: 90.07m | eta: 20.7m step 04530/05568 (81.36%) | loss: 2.390680 | lrm: 0.32 | dt: 1196.27ms | tok/sec: 876,541 | bf16_mfu: 53.84 | total time: 90.09m | eta: 20.7m step 04531/05568 (81.38%) | loss: 2.394577 | lrm: 0.32 | dt: 1196.53ms | tok/sec: 876,349 | bf16_mfu: 53.83 | total time: 90.11m | eta: 20.6m step 04532/05568 (81.39%) | loss: 2.398627 | lrm: 0.32 | dt: 1194.08ms | tok/sec: 878,147 | bf16_mfu: 53.94 | total time: 90.13m | eta: 20.6m step 04533/05568 (81.41%) | loss: 2.390255 | lrm: 0.32 | dt: 1197.29ms | tok/sec: 875,790 | bf16_mfu: 53.80 | total time: 90.15m | eta: 20.6m step 04534/05568 (81.43%) | loss: 2.393301 | lrm: 0.32 | dt: 1189.90ms | tok/sec: 881,229 | bf16_mfu: 54.13 | total time: 90.17m | eta: 20.6m step 04535/05568 (81.45%) | loss: 2.387920 | lrm: 0.32 | dt: 1191.85ms | tok/sec: 879,790 | bf16_mfu: 54.04 | total time: 90.19m | eta: 20.6m step 04536/05568 (81.47%) | loss: 2.400239 | lrm: 0.32 | dt: 1194.35ms | tok/sec: 877,943 | bf16_mfu: 53.93 | total time: 90.21m | eta: 20.5m step 04537/05568 (81.48%) | loss: 2.403955 | lrm: 0.32 | dt: 1191.26ms | tok/sec: 880,224 | bf16_mfu: 54.07 | total time: 90.23m | eta: 20.5m step 04538/05568 (81.50%) | loss: 2.410210 | lrm: 0.32 | dt: 1191.82ms | tok/sec: 879,814 | bf16_mfu: 54.04 | total time: 90.25m | eta: 20.5m step 04539/05568 (81.52%) | loss: 2.405186 | lrm: 0.32 | dt: 1195.88ms | tok/sec: 876,825 | bf16_mfu: 53.86 | total time: 90.27m | eta: 20.5m step 04540/05568 (81.54%) | loss: 2.408180 | lrm: 0.32 | dt: 1198.34ms | tok/sec: 875,022 | bf16_mfu: 53.75 | total time: 90.29m | eta: 20.5m step 04541/05568 (81.56%) | loss: 2.400020 | lrm: 0.32 | dt: 1194.12ms | tok/sec: 878,114 | bf16_mfu: 53.94 | total time: 90.31m | eta: 20.4m step 04542/05568 (81.57%) | loss: 2.409447 | lrm: 0.32 | dt: 1192.62ms | tok/sec: 879,219 | bf16_mfu: 54.01 | total time: 90.33m | eta: 20.4m step 04543/05568 (81.59%) | loss: 2.413166 | lrm: 0.32 | dt: 1194.93ms | tok/sec: 877,518 | bf16_mfu: 53.90 | total time: 90.35m | eta: 20.4m step 04544/05568 (81.61%) | loss: 2.417605 | lrm: 0.32 | dt: 1193.24ms | tok/sec: 878,762 | bf16_mfu: 53.98 | total time: 90.37m | eta: 20.4m step 04545/05568 (81.63%) | loss: 2.420171 | lrm: 0.32 | dt: 1195.69ms | tok/sec: 876,966 | bf16_mfu: 53.87 | total time: 90.39m | eta: 20.4m step 04546/05568 (81.65%) | loss: 2.418971 | lrm: 0.32 | dt: 1195.71ms | tok/sec: 876,949 | bf16_mfu: 53.87 | total time: 90.41m | eta: 20.4m step 04547/05568 (81.66%) | loss: 2.419054 | lrm: 0.32 | dt: 1194.71ms | tok/sec: 877,680 | bf16_mfu: 53.91 | total time: 90.43m | eta: 20.3m step 04548/05568 (81.68%) | loss: 2.412918 | lrm: 0.32 | dt: 1194.12ms | tok/sec: 878,117 | bf16_mfu: 53.94 | total time: 90.45m | eta: 20.3m step 04549/05568 (81.70%) | loss: 2.411600 | lrm: 0.32 | dt: 1195.87ms | tok/sec: 876,831 | bf16_mfu: 53.86 | total time: 90.47m | eta: 20.3m step 04550/05568 (81.72%) | loss: 2.408093 | lrm: 0.32 | dt: 1196.55ms | tok/sec: 876,334 | bf16_mfu: 53.83 | total time: 90.49m | eta: 20.3m step 04551/05568 (81.73%) | loss: 2.415822 | lrm: 0.32 | dt: 1190.32ms | tok/sec: 880,921 | bf16_mfu: 54.11 | total time: 90.51m | eta: 20.3m step 04552/05568 (81.75%) | loss: 2.413659 | lrm: 0.32 | dt: 1200.89ms | tok/sec: 873,168 | bf16_mfu: 53.64 | total time: 90.53m | eta: 20.2m step 04553/05568 (81.77%) | loss: 2.415796 | lrm: 0.32 | dt: 1192.59ms | tok/sec: 879,244 | bf16_mfu: 54.01 | total time: 90.55m | eta: 20.2m step 04554/05568 (81.79%) | loss: 2.416287 | lrm: 0.32 | dt: 1192.37ms | tok/sec: 879,408 | bf16_mfu: 54.02 | total time: 90.57m | eta: 20.2m step 04555/05568 (81.81%) | loss: 2.413420 | lrm: 0.32 | dt: 1196.56ms | tok/sec: 876,322 | bf16_mfu: 53.83 | total time: 90.59m | eta: 20.2m step 04556/05568 (81.82%) | loss: 2.412198 | lrm: 0.32 | dt: 1194.21ms | tok/sec: 878,053 | bf16_mfu: 53.94 | total time: 90.61m | eta: 20.2m step 04557/05568 (81.84%) | loss: 2.426531 | lrm: 0.32 | dt: 1192.99ms | tok/sec: 878,944 | bf16_mfu: 53.99 | total time: 90.63m | eta: 20.1m step 04558/05568 (81.86%) | loss: 2.420782 | lrm: 0.32 | dt: 1193.04ms | tok/sec: 878,910 | bf16_mfu: 53.99 | total time: 90.65m | eta: 20.1m step 04559/05568 (81.88%) | loss: 2.433145 | lrm: 0.31 | dt: 1193.52ms | tok/sec: 878,554 | bf16_mfu: 53.97 | total time: 90.67m | eta: 20.1m step 04560/05568 (81.90%) | loss: 2.428225 | lrm: 0.31 | dt: 1198.19ms | tok/sec: 875,131 | bf16_mfu: 53.76 | total time: 90.69m | eta: 20.1m step 04561/05568 (81.91%) | loss: 2.427974 | lrm: 0.31 | dt: 1195.81ms | tok/sec: 876,874 | bf16_mfu: 53.86 | total time: 90.71m | eta: 20.1m step 04562/05568 (81.93%) | loss: 2.419886 | lrm: 0.31 | dt: 1194.11ms | tok/sec: 878,125 | bf16_mfu: 53.94 | total time: 90.73m | eta: 20.0m step 04563/05568 (81.95%) | loss: 2.418895 | lrm: 0.31 | dt: 1198.44ms | tok/sec: 874,949 | bf16_mfu: 53.75 | total time: 90.75m | eta: 20.0m step 04564/05568 (81.97%) | loss: 2.410740 | lrm: 0.31 | dt: 1193.85ms | tok/sec: 878,311 | bf16_mfu: 53.95 | total time: 90.77m | eta: 20.0m step 04565/05568 (81.99%) | loss: 2.409240 | lrm: 0.31 | dt: 1194.09ms | tok/sec: 878,139 | bf16_mfu: 53.94 | total time: 90.79m | eta: 20.0m step 04566/05568 (82.00%) | loss: 2.408494 | lrm: 0.31 | dt: 1193.86ms | tok/sec: 878,304 | bf16_mfu: 53.95 | total time: 90.81m | eta: 20.0m step 04567/05568 (82.02%) | loss: 2.416724 | lrm: 0.31 | dt: 1199.27ms | tok/sec: 874,348 | bf16_mfu: 53.71 | total time: 90.83m | eta: 19.9m step 04568/05568 (82.04%) | loss: 2.414615 | lrm: 0.31 | dt: 1194.79ms | tok/sec: 877,622 | bf16_mfu: 53.91 | total time: 90.85m | eta: 19.9m step 04569/05568 (82.06%) | loss: 2.401708 | lrm: 0.31 | dt: 1198.29ms | tok/sec: 875,063 | bf16_mfu: 53.75 | total time: 90.87m | eta: 19.9m step 04570/05568 (82.08%) | loss: 2.407197 | lrm: 0.31 | dt: 1197.46ms | tok/sec: 875,670 | bf16_mfu: 53.79 | total time: 90.89m | eta: 19.9m step 04571/05568 (82.09%) | loss: 2.408963 | lrm: 0.31 | dt: 1188.14ms | tok/sec: 882,537 | bf16_mfu: 54.21 | total time: 90.91m | eta: 19.9m step 04572/05568 (82.11%) | loss: 2.411320 | lrm: 0.31 | dt: 1198.24ms | tok/sec: 875,099 | bf16_mfu: 53.76 | total time: 90.93m | eta: 19.8m step 04573/05568 (82.13%) | loss: 2.413740 | lrm: 0.31 | dt: 1196.73ms | tok/sec: 876,201 | bf16_mfu: 53.82 | total time: 90.95m | eta: 19.8m step 04574/05568 (82.15%) | loss: 2.411091 | lrm: 0.31 | dt: 1190.11ms | tok/sec: 881,073 | bf16_mfu: 54.12 | total time: 90.97m | eta: 19.8m step 04575/05568 (82.17%) | loss: 2.408963 | lrm: 0.31 | dt: 1193.65ms | tok/sec: 878,464 | bf16_mfu: 53.96 | total time: 90.99m | eta: 19.8m step 04576/05568 (82.18%) | loss: 2.412772 | lrm: 0.31 | dt: 1198.04ms | tok/sec: 875,241 | bf16_mfu: 53.76 | total time: 91.01m | eta: 19.8m step 04577/05568 (82.20%) | loss: 2.412357 | lrm: 0.31 | dt: 1193.70ms | tok/sec: 878,427 | bf16_mfu: 53.96 | total time: 91.03m | eta: 19.7m step 04578/05568 (82.22%) | loss: 2.412568 | lrm: 0.31 | dt: 1196.28ms | tok/sec: 876,533 | bf16_mfu: 53.84 | total time: 91.05m | eta: 19.7m step 04579/05568 (82.24%) | loss: 2.408507 | lrm: 0.31 | dt: 1194.02ms | tok/sec: 878,188 | bf16_mfu: 53.95 | total time: 91.07m | eta: 19.7m step 04580/05568 (82.26%) | loss: 2.413745 | lrm: 0.31 | dt: 1192.41ms | tok/sec: 879,372 | bf16_mfu: 54.02 | total time: 91.09m | eta: 19.7m step 04581/05568 (82.27%) | loss: 2.411049 | lrm: 0.31 | dt: 1198.14ms | tok/sec: 875,167 | bf16_mfu: 53.76 | total time: 91.11m | eta: 19.7m step 04582/05568 (82.29%) | loss: 2.408889 | lrm: 0.31 | dt: 1198.79ms | tok/sec: 874,696 | bf16_mfu: 53.73 | total time: 91.13m | eta: 19.6m step 04583/05568 (82.31%) | loss: 2.415731 | lrm: 0.31 | dt: 1196.15ms | tok/sec: 876,622 | bf16_mfu: 53.85 | total time: 91.15m | eta: 19.6m step 04584/05568 (82.33%) | loss: 2.414159 | lrm: 0.31 | dt: 1195.20ms | tok/sec: 877,325 | bf16_mfu: 53.89 | total time: 91.17m | eta: 19.6m step 04585/05568 (82.35%) | loss: 2.424975 | lrm: 0.31 | dt: 1195.32ms | tok/sec: 877,233 | bf16_mfu: 53.89 | total time: 91.19m | eta: 19.6m step 04586/05568 (82.36%) | loss: 2.425535 | lrm: 0.31 | dt: 1193.12ms | tok/sec: 878,852 | bf16_mfu: 53.99 | total time: 91.21m | eta: 19.6m step 04587/05568 (82.38%) | loss: 2.422167 | lrm: 0.31 | dt: 1196.05ms | tok/sec: 876,697 | bf16_mfu: 53.85 | total time: 91.23m | eta: 19.5m step 04588/05568 (82.40%) | loss: 2.414404 | lrm: 0.31 | dt: 1196.80ms | tok/sec: 876,146 | bf16_mfu: 53.82 | total time: 91.25m | eta: 19.5m step 04589/05568 (82.42%) | loss: 2.420173 | lrm: 0.31 | dt: 1193.35ms | tok/sec: 878,684 | bf16_mfu: 53.98 | total time: 91.27m | eta: 19.5m step 04590/05568 (82.44%) | loss: 2.422181 | lrm: 0.31 | dt: 1189.50ms | tok/sec: 881,527 | bf16_mfu: 54.15 | total time: 91.29m | eta: 19.5m step 04591/05568 (82.45%) | loss: 2.414722 | lrm: 0.31 | dt: 1196.41ms | tok/sec: 876,438 | bf16_mfu: 53.84 | total time: 91.31m | eta: 19.5m step 04592/05568 (82.47%) | loss: 2.412686 | lrm: 0.31 | dt: 1198.08ms | tok/sec: 875,214 | bf16_mfu: 53.76 | total time: 91.33m | eta: 19.4m step 04593/05568 (82.49%) | loss: 2.412883 | lrm: 0.31 | dt: 1196.12ms | tok/sec: 876,651 | bf16_mfu: 53.85 | total time: 91.35m | eta: 19.4m step 04594/05568 (82.51%) | loss: 2.414568 | lrm: 0.31 | dt: 1200.74ms | tok/sec: 873,272 | bf16_mfu: 53.64 | total time: 91.37m | eta: 19.4m step 04595/05568 (82.53%) | loss: 2.397701 | lrm: 0.31 | dt: 1197.90ms | tok/sec: 875,343 | bf16_mfu: 53.77 | total time: 91.39m | eta: 19.4m step 04596/05568 (82.54%) | loss: 2.400601 | lrm: 0.31 | dt: 1193.08ms | tok/sec: 878,882 | bf16_mfu: 53.99 | total time: 91.41m | eta: 19.4m step 04597/05568 (82.56%) | loss: 2.401477 | lrm: 0.30 | dt: 1195.73ms | tok/sec: 876,936 | bf16_mfu: 53.87 | total time: 91.43m | eta: 19.3m step 04598/05568 (82.58%) | loss: 2.410241 | lrm: 0.30 | dt: 1192.38ms | tok/sec: 879,396 | bf16_mfu: 54.02 | total time: 91.45m | eta: 19.3m step 04599/05568 (82.60%) | loss: 2.407610 | lrm: 0.30 | dt: 1195.31ms | tok/sec: 877,243 | bf16_mfu: 53.89 | total time: 91.47m | eta: 19.3m step 04600/05568 (82.61%) | loss: 2.414261 | lrm: 0.30 | dt: 1193.45ms | tok/sec: 878,609 | bf16_mfu: 53.97 | total time: 91.49m | eta: 19.3m step 04601/05568 (82.63%) | loss: 2.418764 | lrm: 0.30 | dt: 1193.75ms | tok/sec: 878,388 | bf16_mfu: 53.96 | total time: 91.51m | eta: 19.3m step 04602/05568 (82.65%) | loss: 2.432239 | lrm: 0.30 | dt: 1196.56ms | tok/sec: 876,323 | bf16_mfu: 53.83 | total time: 91.53m | eta: 19.2m step 04603/05568 (82.67%) | loss: 2.424255 | lrm: 0.30 | dt: 1196.41ms | tok/sec: 876,432 | bf16_mfu: 53.84 | total time: 91.55m | eta: 19.2m step 04604/05568 (82.69%) | loss: 2.417915 | lrm: 0.30 | dt: 1192.48ms | tok/sec: 879,323 | bf16_mfu: 54.01 | total time: 91.57m | eta: 19.2m step 04605/05568 (82.70%) | loss: 2.417832 | lrm: 0.30 | dt: 1194.06ms | tok/sec: 878,157 | bf16_mfu: 53.94 | total time: 91.59m | eta: 19.2m step 04606/05568 (82.72%) | loss: 2.406264 | lrm: 0.30 | dt: 1195.84ms | tok/sec: 876,851 | bf16_mfu: 53.86 | total time: 91.61m | eta: 19.2m step 04607/05568 (82.74%) | loss: 2.408485 | lrm: 0.30 | dt: 1196.08ms | tok/sec: 876,680 | bf16_mfu: 53.85 | total time: 91.62m | eta: 19.1m step 04608/05568 (82.76%) | loss: 2.403433 | lrm: 0.30 | dt: 1193.24ms | tok/sec: 878,767 | bf16_mfu: 53.98 | total time: 91.64m | eta: 19.1m step 04609/05568 (82.78%) | loss: 2.395671 | lrm: 0.30 | dt: 1192.86ms | tok/sec: 879,043 | bf16_mfu: 54.00 | total time: 91.66m | eta: 19.1m step 04610/05568 (82.79%) | loss: 2.397275 | lrm: 0.30 | dt: 1195.31ms | tok/sec: 877,245 | bf16_mfu: 53.89 | total time: 91.68m | eta: 19.1m step 04611/05568 (82.81%) | loss: 2.393747 | lrm: 0.30 | dt: 1196.57ms | tok/sec: 876,316 | bf16_mfu: 53.83 | total time: 91.70m | eta: 19.1m step 04612/05568 (82.83%) | loss: 2.399514 | lrm: 0.30 | dt: 1194.08ms | tok/sec: 878,146 | bf16_mfu: 53.94 | total time: 91.72m | eta: 19.0m step 04613/05568 (82.85%) | loss: 2.404186 | lrm: 0.30 | dt: 1193.22ms | tok/sec: 878,777 | bf16_mfu: 53.98 | total time: 91.74m | eta: 19.0m step 04614/05568 (82.87%) | loss: 2.407042 | lrm: 0.30 | dt: 1195.16ms | tok/sec: 877,351 | bf16_mfu: 53.89 | total time: 91.76m | eta: 19.0m step 04615/05568 (82.88%) | loss: 2.414724 | lrm: 0.30 | dt: 1193.44ms | tok/sec: 878,615 | bf16_mfu: 53.97 | total time: 91.78m | eta: 19.0m step 04616/05568 (82.90%) | loss: 2.405647 | lrm: 0.30 | dt: 1196.11ms | tok/sec: 876,651 | bf16_mfu: 53.85 | total time: 91.80m | eta: 19.0m step 04617/05568 (82.92%) | loss: 2.405441 | lrm: 0.30 | dt: 1196.43ms | tok/sec: 876,420 | bf16_mfu: 53.84 | total time: 91.82m | eta: 18.9m step 04618/05568 (82.94%) | loss: 2.408658 | lrm: 0.30 | dt: 1193.56ms | tok/sec: 878,524 | bf16_mfu: 53.97 | total time: 91.84m | eta: 18.9m step 04619/05568 (82.96%) | loss: 2.408886 | lrm: 0.30 | dt: 1192.77ms | tok/sec: 879,110 | bf16_mfu: 54.00 | total time: 91.86m | eta: 18.9m step 04620/05568 (82.97%) | loss: 2.414010 | lrm: 0.30 | dt: 1191.91ms | tok/sec: 879,747 | bf16_mfu: 54.04 | total time: 91.88m | eta: 18.9m step 04621/05568 (82.99%) | loss: 2.407089 | lrm: 0.30 | dt: 1197.00ms | tok/sec: 876,004 | bf16_mfu: 53.81 | total time: 91.90m | eta: 18.9m step 04622/05568 (83.01%) | loss: 2.408785 | lrm: 0.30 | dt: 1196.81ms | tok/sec: 876,139 | bf16_mfu: 53.82 | total time: 91.92m | eta: 18.8m step 04623/05568 (83.03%) | loss: 2.415927 | lrm: 0.30 | dt: 1196.58ms | tok/sec: 876,311 | bf16_mfu: 53.83 | total time: 91.94m | eta: 18.8m step 04624/05568 (83.05%) | loss: 2.412792 | lrm: 0.30 | dt: 1191.75ms | tok/sec: 879,865 | bf16_mfu: 54.05 | total time: 91.96m | eta: 18.8m step 04625/05568 (83.06%) | loss: 2.413743 | lrm: 0.30 | dt: 1199.06ms | tok/sec: 874,500 | bf16_mfu: 53.72 | total time: 91.98m | eta: 18.8m step 04626/05568 (83.08%) | loss: 2.405513 | lrm: 0.30 | dt: 1197.08ms | tok/sec: 875,941 | bf16_mfu: 53.81 | total time: 92.00m | eta: 18.8m step 04627/05568 (83.10%) | loss: 2.399278 | lrm: 0.30 | dt: 1196.95ms | tok/sec: 876,041 | bf16_mfu: 53.81 | total time: 92.02m | eta: 18.7m step 04628/05568 (83.12%) | loss: 2.401395 | lrm: 0.30 | dt: 1198.70ms | tok/sec: 874,760 | bf16_mfu: 53.73 | total time: 92.04m | eta: 18.7m step 04629/05568 (83.14%) | loss: 2.396721 | lrm: 0.30 | dt: 1193.57ms | tok/sec: 878,522 | bf16_mfu: 53.97 | total time: 92.06m | eta: 18.7m step 04630/05568 (83.15%) | loss: 2.401075 | lrm: 0.30 | dt: 1193.35ms | tok/sec: 878,680 | bf16_mfu: 53.98 | total time: 92.08m | eta: 18.7m step 04631/05568 (83.17%) | loss: 2.391814 | lrm: 0.30 | dt: 1192.01ms | tok/sec: 879,666 | bf16_mfu: 54.04 | total time: 92.10m | eta: 18.7m step 04632/05568 (83.19%) | loss: 2.388325 | lrm: 0.30 | dt: 1196.62ms | tok/sec: 876,284 | bf16_mfu: 53.83 | total time: 92.12m | eta: 18.6m step 04633/05568 (83.21%) | loss: 2.369501 | lrm: 0.30 | dt: 1192.79ms | tok/sec: 879,091 | bf16_mfu: 54.00 | total time: 92.14m | eta: 18.6m step 04634/05568 (83.23%) | loss: 2.373086 | lrm: 0.30 | dt: 1195.42ms | tok/sec: 877,159 | bf16_mfu: 53.88 | total time: 92.16m | eta: 18.6m step 04635/05568 (83.24%) | loss: 2.375864 | lrm: 0.29 | dt: 1194.05ms | tok/sec: 878,168 | bf16_mfu: 53.94 | total time: 92.18m | eta: 18.6m step 04636/05568 (83.26%) | loss: 2.374769 | lrm: 0.29 | dt: 1198.67ms | tok/sec: 874,782 | bf16_mfu: 53.74 | total time: 92.20m | eta: 18.6m step 04637/05568 (83.28%) | loss: 2.388468 | lrm: 0.29 | dt: 1194.13ms | tok/sec: 878,112 | bf16_mfu: 53.94 | total time: 92.22m | eta: 18.5m step 04638/05568 (83.30%) | loss: 2.390975 | lrm: 0.29 | dt: 1195.01ms | tok/sec: 877,460 | bf16_mfu: 53.90 | total time: 92.24m | eta: 18.5m step 04639/05568 (83.32%) | loss: 2.393429 | lrm: 0.29 | dt: 1196.40ms | tok/sec: 876,440 | bf16_mfu: 53.84 | total time: 92.26m | eta: 18.5m step 04640/05568 (83.33%) | loss: 2.397272 | lrm: 0.29 | dt: 1193.65ms | tok/sec: 878,458 | bf16_mfu: 53.96 | total time: 92.28m | eta: 18.5m step 04641/05568 (83.35%) | loss: 2.410664 | lrm: 0.29 | dt: 1195.50ms | tok/sec: 877,100 | bf16_mfu: 53.88 | total time: 92.30m | eta: 18.5m step 04642/05568 (83.37%) | loss: 2.398275 | lrm: 0.29 | dt: 1195.50ms | tok/sec: 877,101 | bf16_mfu: 53.88 | total time: 92.32m | eta: 18.4m step 04643/05568 (83.39%) | loss: 2.405123 | lrm: 0.29 | dt: 1195.25ms | tok/sec: 877,282 | bf16_mfu: 53.89 | total time: 92.34m | eta: 18.4m step 04644/05568 (83.41%) | loss: 2.423064 | lrm: 0.29 | dt: 1193.25ms | tok/sec: 878,757 | bf16_mfu: 53.98 | total time: 92.36m | eta: 18.4m step 04645/05568 (83.42%) | loss: 2.423852 | lrm: 0.29 | dt: 1197.20ms | tok/sec: 875,857 | bf16_mfu: 53.80 | total time: 92.38m | eta: 18.4m step 04646/05568 (83.44%) | loss: 2.422553 | lrm: 0.29 | dt: 1198.70ms | tok/sec: 874,762 | bf16_mfu: 53.73 | total time: 92.40m | eta: 18.4m step 04647/05568 (83.46%) | loss: 2.416051 | lrm: 0.29 | dt: 1192.58ms | tok/sec: 879,250 | bf16_mfu: 54.01 | total time: 92.42m | eta: 18.3m step 04648/05568 (83.48%) | loss: 2.410516 | lrm: 0.29 | dt: 1198.29ms | tok/sec: 875,060 | bf16_mfu: 53.75 | total time: 92.44m | eta: 18.3m step 04649/05568 (83.49%) | loss: 2.409031 | lrm: 0.29 | dt: 1198.24ms | tok/sec: 875,094 | bf16_mfu: 53.76 | total time: 92.46m | eta: 18.3m step 04650/05568 (83.51%) | loss: 2.410227 | lrm: 0.29 | dt: 1192.58ms | tok/sec: 879,253 | bf16_mfu: 54.01 | total time: 92.48m | eta: 18.3m step 04651/05568 (83.53%) | loss: 2.416021 | lrm: 0.29 | dt: 1197.64ms | tok/sec: 875,538 | bf16_mfu: 53.78 | total time: 92.50m | eta: 18.3m step 04652/05568 (83.55%) | loss: 2.409305 | lrm: 0.29 | dt: 1190.95ms | tok/sec: 880,450 | bf16_mfu: 54.08 | total time: 92.52m | eta: 18.2m step 04653/05568 (83.57%) | loss: 2.411923 | lrm: 0.29 | dt: 1195.03ms | tok/sec: 877,448 | bf16_mfu: 53.90 | total time: 92.54m | eta: 18.2m step 04654/05568 (83.58%) | loss: 2.415623 | lrm: 0.29 | dt: 1192.71ms | tok/sec: 879,156 | bf16_mfu: 54.00 | total time: 92.56m | eta: 18.2m step 04655/05568 (83.60%) | loss: 2.417075 | lrm: 0.29 | dt: 1198.14ms | tok/sec: 875,167 | bf16_mfu: 53.76 | total time: 92.58m | eta: 18.2m step 04656/05568 (83.62%) | loss: 2.423053 | lrm: 0.29 | dt: 1196.23ms | tok/sec: 876,564 | bf16_mfu: 53.85 | total time: 92.60m | eta: 18.2m step 04657/05568 (83.64%) | loss: 2.418595 | lrm: 0.29 | dt: 1196.28ms | tok/sec: 876,530 | bf16_mfu: 53.84 | total time: 92.62m | eta: 18.1m step 04658/05568 (83.66%) | loss: 2.414379 | lrm: 0.29 | dt: 1198.14ms | tok/sec: 875,171 | bf16_mfu: 53.76 | total time: 92.64m | eta: 18.1m step 04659/05568 (83.67%) | loss: 2.412591 | lrm: 0.29 | dt: 1196.02ms | tok/sec: 876,717 | bf16_mfu: 53.85 | total time: 92.66m | eta: 18.1m step 04660/05568 (83.69%) | loss: 2.415162 | lrm: 0.29 | dt: 1195.67ms | tok/sec: 876,979 | bf16_mfu: 53.87 | total time: 92.68m | eta: 18.1m step 04661/05568 (83.71%) | loss: 2.413439 | lrm: 0.29 | dt: 1197.39ms | tok/sec: 875,715 | bf16_mfu: 53.79 | total time: 92.70m | eta: 18.1m step 04662/05568 (83.73%) | loss: 2.410908 | lrm: 0.29 | dt: 1191.65ms | tok/sec: 879,935 | bf16_mfu: 54.05 | total time: 92.72m | eta: 18.0m step 04663/05568 (83.75%) | loss: 2.405066 | lrm: 0.29 | dt: 1194.93ms | tok/sec: 877,518 | bf16_mfu: 53.90 | total time: 92.74m | eta: 18.0m step 04664/05568 (83.76%) | loss: 2.418454 | lrm: 0.29 | dt: 1199.77ms | tok/sec: 873,978 | bf16_mfu: 53.69 | total time: 92.76m | eta: 18.0m step 04665/05568 (83.78%) | loss: 2.412555 | lrm: 0.29 | dt: 1195.52ms | tok/sec: 877,085 | bf16_mfu: 53.88 | total time: 92.78m | eta: 18.0m step 04666/05568 (83.80%) | loss: 2.414067 | lrm: 0.29 | dt: 1192.82ms | tok/sec: 879,074 | bf16_mfu: 54.00 | total time: 92.80m | eta: 18.0m step 04667/05568 (83.82%) | loss: 2.413849 | lrm: 0.29 | dt: 1193.09ms | tok/sec: 878,874 | bf16_mfu: 53.99 | total time: 92.82m | eta: 17.9m step 04668/05568 (83.84%) | loss: 2.417554 | lrm: 0.29 | dt: 1191.23ms | tok/sec: 880,249 | bf16_mfu: 54.07 | total time: 92.84m | eta: 17.9m step 04669/05568 (83.85%) | loss: 2.412511 | lrm: 0.29 | dt: 1198.36ms | tok/sec: 875,005 | bf16_mfu: 53.75 | total time: 92.86m | eta: 17.9m step 04670/05568 (83.87%) | loss: 2.409565 | lrm: 0.29 | dt: 1190.40ms | tok/sec: 880,861 | bf16_mfu: 54.11 | total time: 92.88m | eta: 17.9m step 04671/05568 (83.89%) | loss: 2.404958 | lrm: 0.29 | dt: 1197.23ms | tok/sec: 875,831 | bf16_mfu: 53.80 | total time: 92.90m | eta: 17.9m step 04672/05568 (83.91%) | loss: 2.413674 | lrm: 0.29 | dt: 1197.56ms | tok/sec: 875,595 | bf16_mfu: 53.79 | total time: 92.92m | eta: 17.8m step 04673/05568 (83.93%) | loss: 2.417279 | lrm: 0.28 | dt: 1197.98ms | tok/sec: 875,286 | bf16_mfu: 53.77 | total time: 92.94m | eta: 17.8m step 04674/05568 (83.94%) | loss: 2.431709 | lrm: 0.28 | dt: 1194.42ms | tok/sec: 877,896 | bf16_mfu: 53.93 | total time: 92.96m | eta: 17.8m step 04675/05568 (83.96%) | loss: 2.439215 | lrm: 0.28 | dt: 1192.90ms | tok/sec: 879,017 | bf16_mfu: 54.00 | total time: 92.98m | eta: 17.8m step 04676/05568 (83.98%) | loss: 2.430945 | lrm: 0.28 | dt: 1194.28ms | tok/sec: 877,996 | bf16_mfu: 53.93 | total time: 93.00m | eta: 17.8m step 04677/05568 (84.00%) | loss: 2.428452 | lrm: 0.28 | dt: 1192.27ms | tok/sec: 879,476 | bf16_mfu: 54.02 | total time: 93.02m | eta: 17.7m step 04678/05568 (84.02%) | loss: 2.429366 | lrm: 0.28 | dt: 1192.17ms | tok/sec: 879,549 | bf16_mfu: 54.03 | total time: 93.04m | eta: 17.7m step 04679/05568 (84.03%) | loss: 2.426137 | lrm: 0.28 | dt: 1196.36ms | tok/sec: 876,475 | bf16_mfu: 53.84 | total time: 93.06m | eta: 17.7m step 04680/05568 (84.05%) | loss: 2.422391 | lrm: 0.28 | dt: 1196.60ms | tok/sec: 876,299 | bf16_mfu: 53.83 | total time: 93.08m | eta: 17.7m step 04681/05568 (84.07%) | loss: 2.411090 | lrm: 0.28 | dt: 1195.83ms | tok/sec: 876,860 | bf16_mfu: 53.86 | total time: 93.10m | eta: 17.7m step 04682/05568 (84.09%) | loss: 2.417218 | lrm: 0.28 | dt: 1189.07ms | tok/sec: 881,844 | bf16_mfu: 54.17 | total time: 93.12m | eta: 17.6m step 04683/05568 (84.11%) | loss: 2.416880 | lrm: 0.28 | dt: 1195.07ms | tok/sec: 877,418 | bf16_mfu: 53.90 | total time: 93.14m | eta: 17.6m step 04684/05568 (84.12%) | loss: 2.423176 | lrm: 0.28 | dt: 1192.45ms | tok/sec: 879,347 | bf16_mfu: 54.02 | total time: 93.16m | eta: 17.6m step 04685/05568 (84.14%) | loss: 2.427311 | lrm: 0.28 | dt: 1193.80ms | tok/sec: 878,352 | bf16_mfu: 53.96 | total time: 93.18m | eta: 17.6m step 04686/05568 (84.16%) | loss: 2.429766 | lrm: 0.28 | dt: 1192.82ms | tok/sec: 879,073 | bf16_mfu: 54.00 | total time: 93.20m | eta: 17.6m step 04687/05568 (84.18%) | loss: 2.433666 | lrm: 0.28 | dt: 1196.88ms | tok/sec: 876,088 | bf16_mfu: 53.82 | total time: 93.22m | eta: 17.5m step 04688/05568 (84.20%) | loss: 2.425856 | lrm: 0.28 | dt: 1194.42ms | tok/sec: 877,895 | bf16_mfu: 53.93 | total time: 93.24m | eta: 17.5m step 04689/05568 (84.21%) | loss: 2.422127 | lrm: 0.28 | dt: 1193.42ms | tok/sec: 878,633 | bf16_mfu: 53.97 | total time: 93.26m | eta: 17.5m step 04690/05568 (84.23%) | loss: 2.420387 | lrm: 0.28 | dt: 1193.52ms | tok/sec: 878,560 | bf16_mfu: 53.97 | total time: 93.28m | eta: 17.5m step 04691/05568 (84.25%) | loss: 2.427652 | lrm: 0.28 | dt: 1192.13ms | tok/sec: 879,581 | bf16_mfu: 54.03 | total time: 93.30m | eta: 17.5m step 04692/05568 (84.27%) | loss: 2.435170 | lrm: 0.28 | dt: 1194.99ms | tok/sec: 877,473 | bf16_mfu: 53.90 | total time: 93.32m | eta: 17.4m step 04693/05568 (84.29%) | loss: 2.443216 | lrm: 0.28 | dt: 1197.05ms | tok/sec: 875,970 | bf16_mfu: 53.81 | total time: 93.34m | eta: 17.4m step 04694/05568 (84.30%) | loss: 2.444539 | lrm: 0.28 | dt: 1198.83ms | tok/sec: 874,667 | bf16_mfu: 53.73 | total time: 93.36m | eta: 17.4m step 04695/05568 (84.32%) | loss: 2.446304 | lrm: 0.28 | dt: 1194.27ms | tok/sec: 878,004 | bf16_mfu: 53.93 | total time: 93.38m | eta: 17.4m step 04696/05568 (84.34%) | loss: 2.443342 | lrm: 0.28 | dt: 1191.69ms | tok/sec: 879,909 | bf16_mfu: 54.05 | total time: 93.40m | eta: 17.4m step 04697/05568 (84.36%) | loss: 2.441912 | lrm: 0.28 | dt: 1195.09ms | tok/sec: 877,401 | bf16_mfu: 53.90 | total time: 93.42m | eta: 17.3m step 04698/05568 (84.38%) | loss: 2.437001 | lrm: 0.28 | dt: 1196.32ms | tok/sec: 876,503 | bf16_mfu: 53.84 | total time: 93.44m | eta: 17.3m step 04699/05568 (84.39%) | loss: 2.432027 | lrm: 0.28 | dt: 1193.05ms | tok/sec: 878,904 | bf16_mfu: 53.99 | total time: 93.46m | eta: 17.3m step 04700/05568 (84.41%) | loss: 2.429611 | lrm: 0.28 | dt: 1193.44ms | tok/sec: 878,613 | bf16_mfu: 53.97 | total time: 93.48m | eta: 17.3m step 04701/05568 (84.43%) | loss: 2.427966 | lrm: 0.28 | dt: 1194.07ms | tok/sec: 878,154 | bf16_mfu: 53.94 | total time: 93.50m | eta: 17.3m step 04702/05568 (84.45%) | loss: 2.418256 | lrm: 0.28 | dt: 1198.22ms | tok/sec: 875,112 | bf16_mfu: 53.76 | total time: 93.52m | eta: 17.2m step 04703/05568 (84.46%) | loss: 2.428955 | lrm: 0.28 | dt: 1197.65ms | tok/sec: 875,529 | bf16_mfu: 53.78 | total time: 93.54m | eta: 17.2m step 04704/05568 (84.48%) | loss: 2.421488 | lrm: 0.28 | dt: 1199.16ms | tok/sec: 874,426 | bf16_mfu: 53.71 | total time: 93.56m | eta: 17.2m step 04705/05568 (84.50%) | loss: 2.415371 | lrm: 0.28 | dt: 1195.27ms | tok/sec: 877,268 | bf16_mfu: 53.89 | total time: 93.58m | eta: 17.2m step 04706/05568 (84.52%) | loss: 2.417991 | lrm: 0.28 | dt: 1194.64ms | tok/sec: 877,732 | bf16_mfu: 53.92 | total time: 93.60m | eta: 17.2m step 04707/05568 (84.54%) | loss: 2.425098 | lrm: 0.28 | dt: 1194.48ms | tok/sec: 877,851 | bf16_mfu: 53.92 | total time: 93.62m | eta: 17.1m step 04708/05568 (84.55%) | loss: 2.421251 | lrm: 0.28 | dt: 1191.47ms | tok/sec: 880,070 | bf16_mfu: 54.06 | total time: 93.64m | eta: 17.1m step 04709/05568 (84.57%) | loss: 2.426709 | lrm: 0.28 | dt: 1187.65ms | tok/sec: 882,901 | bf16_mfu: 54.23 | total time: 93.66m | eta: 17.1m step 04710/05568 (84.59%) | loss: 2.423667 | lrm: 0.28 | dt: 1198.27ms | tok/sec: 875,073 | bf16_mfu: 53.75 | total time: 93.68m | eta: 17.1m step 04711/05568 (84.61%) | loss: 2.423442 | lrm: 0.27 | dt: 1193.45ms | tok/sec: 878,612 | bf16_mfu: 53.97 | total time: 93.70m | eta: 17.1m step 04712/05568 (84.63%) | loss: 2.417242 | lrm: 0.27 | dt: 1191.41ms | tok/sec: 880,116 | bf16_mfu: 54.06 | total time: 93.72m | eta: 17.0m step 04713/05568 (84.64%) | loss: 2.415246 | lrm: 0.27 | dt: 1193.56ms | tok/sec: 878,531 | bf16_mfu: 53.97 | total time: 93.74m | eta: 17.0m step 04714/05568 (84.66%) | loss: 2.419365 | lrm: 0.27 | dt: 1196.68ms | tok/sec: 876,234 | bf16_mfu: 53.83 | total time: 93.76m | eta: 17.0m step 04715/05568 (84.68%) | loss: 2.417163 | lrm: 0.27 | dt: 1194.84ms | tok/sec: 877,587 | bf16_mfu: 53.91 | total time: 93.78m | eta: 17.0m step 04716/05568 (84.70%) | loss: 2.405846 | lrm: 0.27 | dt: 1196.19ms | tok/sec: 876,596 | bf16_mfu: 53.85 | total time: 93.80m | eta: 17.0m step 04717/05568 (84.72%) | loss: 2.413469 | lrm: 0.27 | dt: 1196.32ms | tok/sec: 876,504 | bf16_mfu: 53.84 | total time: 93.82m | eta: 16.9m step 04718/05568 (84.73%) | loss: 2.423436 | lrm: 0.27 | dt: 1197.10ms | tok/sec: 875,926 | bf16_mfu: 53.81 | total time: 93.84m | eta: 16.9m step 04719/05568 (84.75%) | loss: 2.411680 | lrm: 0.27 | dt: 1194.89ms | tok/sec: 877,553 | bf16_mfu: 53.91 | total time: 93.86m | eta: 16.9m step 04720/05568 (84.77%) | loss: 2.418380 | lrm: 0.27 | dt: 1198.99ms | tok/sec: 874,549 | bf16_mfu: 53.72 | total time: 93.88m | eta: 16.9m step 04721/05568 (84.79%) | loss: 2.420335 | lrm: 0.27 | dt: 1190.63ms | tok/sec: 880,691 | bf16_mfu: 54.10 | total time: 93.90m | eta: 16.9m step 04722/05568 (84.81%) | loss: 2.422001 | lrm: 0.27 | dt: 1193.50ms | tok/sec: 878,570 | bf16_mfu: 53.97 | total time: 93.92m | eta: 16.8m step 04723/05568 (84.82%) | loss: 2.420149 | lrm: 0.27 | dt: 1191.68ms | tok/sec: 879,910 | bf16_mfu: 54.05 | total time: 93.94m | eta: 16.8m step 04724/05568 (84.84%) | loss: 2.413773 | lrm: 0.27 | dt: 1197.64ms | tok/sec: 875,537 | bf16_mfu: 53.78 | total time: 93.96m | eta: 16.8m step 04725/05568 (84.86%) | loss: 2.399665 | lrm: 0.27 | dt: 1190.44ms | tok/sec: 880,831 | bf16_mfu: 54.11 | total time: 93.97m | eta: 16.8m step 04726/05568 (84.88%) | loss: 2.402944 | lrm: 0.27 | dt: 1197.11ms | tok/sec: 875,919 | bf16_mfu: 53.81 | total time: 93.99m | eta: 16.8m step 04727/05568 (84.90%) | loss: 2.401278 | lrm: 0.27 | dt: 1192.53ms | tok/sec: 879,284 | bf16_mfu: 54.01 | total time: 94.01m | eta: 16.7m step 04728/05568 (84.91%) | loss: 2.399395 | lrm: 0.27 | dt: 1197.03ms | tok/sec: 875,980 | bf16_mfu: 53.81 | total time: 94.03m | eta: 16.7m step 04729/05568 (84.93%) | loss: 2.407571 | lrm: 0.27 | dt: 1196.28ms | tok/sec: 876,531 | bf16_mfu: 53.84 | total time: 94.05m | eta: 16.7m step 04730/05568 (84.95%) | loss: 2.397570 | lrm: 0.27 | dt: 1190.40ms | tok/sec: 880,861 | bf16_mfu: 54.11 | total time: 94.07m | eta: 16.7m step 04731/05568 (84.97%) | loss: 2.406973 | lrm: 0.27 | dt: 1199.36ms | tok/sec: 874,282 | bf16_mfu: 53.71 | total time: 94.09m | eta: 16.7m step 04732/05568 (84.99%) | loss: 2.415452 | lrm: 0.27 | dt: 1192.20ms | tok/sec: 879,532 | bf16_mfu: 54.03 | total time: 94.11m | eta: 16.6m step 04733/05568 (85.00%) | loss: 2.419649 | lrm: 0.27 | dt: 1191.98ms | tok/sec: 879,689 | bf16_mfu: 54.04 | total time: 94.13m | eta: 16.6m step 04734/05568 (85.02%) | loss: 2.416220 | lrm: 0.27 | dt: 1192.33ms | tok/sec: 879,435 | bf16_mfu: 54.02 | total time: 94.15m | eta: 16.6m step 04735/05568 (85.04%) | loss: 2.410586 | lrm: 0.27 | dt: 1196.66ms | tok/sec: 876,253 | bf16_mfu: 53.83 | total time: 94.17m | eta: 16.6m step 04736/05568 (85.06%) | loss: 2.411778 | lrm: 0.27 | dt: 1195.35ms | tok/sec: 877,208 | bf16_mfu: 53.88 | total time: 94.19m | eta: 16.6m step 04737/05568 (85.08%) | loss: 2.420946 | lrm: 0.27 | dt: 1196.56ms | tok/sec: 876,324 | bf16_mfu: 53.83 | total time: 94.21m | eta: 16.5m step 04738/05568 (85.09%) | loss: 2.419360 | lrm: 0.27 | dt: 1197.81ms | tok/sec: 875,410 | bf16_mfu: 53.77 | total time: 94.23m | eta: 16.5m step 04739/05568 (85.11%) | loss: 2.410792 | lrm: 0.27 | dt: 1189.43ms | tok/sec: 881,578 | bf16_mfu: 54.15 | total time: 94.25m | eta: 16.5m step 04740/05568 (85.13%) | loss: 2.416496 | lrm: 0.27 | dt: 1194.80ms | tok/sec: 877,615 | bf16_mfu: 53.91 | total time: 94.27m | eta: 16.5m step 04741/05568 (85.15%) | loss: 2.415333 | lrm: 0.27 | dt: 1198.24ms | tok/sec: 875,093 | bf16_mfu: 53.75 | total time: 94.29m | eta: 16.5m step 04742/05568 (85.17%) | loss: 2.414902 | lrm: 0.27 | dt: 1191.49ms | tok/sec: 880,055 | bf16_mfu: 54.06 | total time: 94.31m | eta: 16.4m step 04743/05568 (85.18%) | loss: 2.399608 | lrm: 0.27 | dt: 1193.44ms | tok/sec: 878,618 | bf16_mfu: 53.97 | total time: 94.33m | eta: 16.4m step 04744/05568 (85.20%) | loss: 2.392032 | lrm: 0.27 | dt: 1201.20ms | tok/sec: 872,936 | bf16_mfu: 53.62 | total time: 94.35m | eta: 16.4m step 04745/05568 (85.22%) | loss: 2.399589 | lrm: 0.27 | dt: 1192.81ms | tok/sec: 879,081 | bf16_mfu: 54.00 | total time: 94.37m | eta: 16.4m step 04746/05568 (85.24%) | loss: 2.409259 | lrm: 0.27 | dt: 1195.57ms | tok/sec: 877,048 | bf16_mfu: 53.88 | total time: 94.39m | eta: 16.4m step 04747/05568 (85.26%) | loss: 2.411458 | lrm: 0.27 | dt: 1195.12ms | tok/sec: 877,380 | bf16_mfu: 53.90 | total time: 94.41m | eta: 16.3m step 04748/05568 (85.27%) | loss: 2.412621 | lrm: 0.27 | dt: 1191.31ms | tok/sec: 880,190 | bf16_mfu: 54.07 | total time: 94.43m | eta: 16.3m step 04749/05568 (85.29%) | loss: 2.417584 | lrm: 0.26 | dt: 1196.62ms | tok/sec: 876,284 | bf16_mfu: 53.83 | total time: 94.45m | eta: 16.3m step:4750/5568 val_bpb:0.731419 val_time:3.89s step 04750/05568 (85.31%) | loss: 2.415825 | lrm: 0.26 | dt: 1193.53ms | tok/sec: 878,547 | bf16_mfu: 53.97 | total time: 94.47m | eta: 16.3m step 04751/05568 (85.33%) | loss: 2.421293 | lrm: 0.26 | dt: 1192.19ms | tok/sec: 879,535 | bf16_mfu: 54.03 | total time: 94.49m | eta: 16.3m step 04752/05568 (85.34%) | loss: 2.419300 | lrm: 0.26 | dt: 1199.72ms | tok/sec: 874,014 | bf16_mfu: 53.69 | total time: 94.51m | eta: 16.2m step 04753/05568 (85.36%) | loss: 2.424289 | lrm: 0.26 | dt: 1198.60ms | tok/sec: 874,832 | bf16_mfu: 53.74 | total time: 94.53m | eta: 16.2m step 04754/05568 (85.38%) | loss: 2.414982 | lrm: 0.26 | dt: 1193.36ms | tok/sec: 878,673 | bf16_mfu: 53.97 | total time: 94.55m | eta: 16.2m step 04755/05568 (85.40%) | loss: 2.417008 | lrm: 0.26 | dt: 1193.30ms | tok/sec: 878,717 | bf16_mfu: 53.98 | total time: 94.57m | eta: 16.2m step 04756/05568 (85.42%) | loss: 2.411502 | lrm: 0.26 | dt: 1194.16ms | tok/sec: 878,089 | bf16_mfu: 53.94 | total time: 94.59m | eta: 16.2m step 04757/05568 (85.43%) | loss: 2.416081 | lrm: 0.26 | dt: 1200.39ms | tok/sec: 873,526 | bf16_mfu: 53.66 | total time: 94.61m | eta: 16.1m step 04758/05568 (85.45%) | loss: 2.427542 | lrm: 0.26 | dt: 1195.54ms | tok/sec: 877,072 | bf16_mfu: 53.88 | total time: 94.63m | eta: 16.1m step 04759/05568 (85.47%) | loss: 2.432080 | lrm: 0.26 | dt: 1193.87ms | tok/sec: 878,301 | bf16_mfu: 53.95 | total time: 94.65m | eta: 16.1m step 04760/05568 (85.49%) | loss: 2.424919 | lrm: 0.26 | dt: 1194.42ms | tok/sec: 877,897 | bf16_mfu: 53.93 | total time: 94.67m | eta: 16.1m step 04761/05568 (85.51%) | loss: 2.427908 | lrm: 0.26 | dt: 1192.71ms | tok/sec: 879,153 | bf16_mfu: 54.00 | total time: 94.69m | eta: 16.1m step 04762/05568 (85.52%) | loss: 2.433238 | lrm: 0.26 | dt: 1197.09ms | tok/sec: 875,934 | bf16_mfu: 53.81 | total time: 94.71m | eta: 16.0m step 04763/05568 (85.54%) | loss: 2.430526 | lrm: 0.26 | dt: 1195.26ms | tok/sec: 877,277 | bf16_mfu: 53.89 | total time: 94.73m | eta: 16.0m step 04764/05568 (85.56%) | loss: 2.415819 | lrm: 0.26 | dt: 1194.46ms | tok/sec: 877,866 | bf16_mfu: 53.93 | total time: 94.75m | eta: 16.0m step 04765/05568 (85.58%) | loss: 2.419493 | lrm: 0.26 | dt: 1191.51ms | tok/sec: 880,036 | bf16_mfu: 54.06 | total time: 94.77m | eta: 16.0m step 04766/05568 (85.60%) | loss: 2.415020 | lrm: 0.26 | dt: 1196.85ms | tok/sec: 876,114 | bf16_mfu: 53.82 | total time: 94.79m | eta: 16.0m step 04767/05568 (85.61%) | loss: 2.423568 | lrm: 0.26 | dt: 1194.96ms | tok/sec: 877,499 | bf16_mfu: 53.90 | total time: 94.81m | eta: 15.9m step 04768/05568 (85.63%) | loss: 2.421408 | lrm: 0.26 | dt: 1192.47ms | tok/sec: 879,331 | bf16_mfu: 54.02 | total time: 94.83m | eta: 15.9m step 04769/05568 (85.65%) | loss: 2.416945 | lrm: 0.26 | dt: 1199.14ms | tok/sec: 874,439 | bf16_mfu: 53.71 | total time: 94.85m | eta: 15.9m step 04770/05568 (85.67%) | loss: 2.424602 | lrm: 0.26 | dt: 1193.53ms | tok/sec: 878,549 | bf16_mfu: 53.97 | total time: 94.87m | eta: 15.9m step 04771/05568 (85.69%) | loss: 2.422985 | lrm: 0.26 | dt: 1195.39ms | tok/sec: 877,181 | bf16_mfu: 53.88 | total time: 94.89m | eta: 15.9m step 04772/05568 (85.70%) | loss: 2.432811 | lrm: 0.26 | dt: 1192.42ms | tok/sec: 879,368 | bf16_mfu: 54.02 | total time: 94.91m | eta: 15.8m step 04773/05568 (85.72%) | loss: 2.433803 | lrm: 0.26 | dt: 1192.76ms | tok/sec: 879,117 | bf16_mfu: 54.00 | total time: 94.93m | eta: 15.8m step 04774/05568 (85.74%) | loss: 2.423071 | lrm: 0.26 | dt: 1195.19ms | tok/sec: 877,327 | bf16_mfu: 53.89 | total time: 94.95m | eta: 15.8m step 04775/05568 (85.76%) | loss: 2.437175 | lrm: 0.26 | dt: 1192.49ms | tok/sec: 879,314 | bf16_mfu: 54.01 | total time: 94.97m | eta: 15.8m step 04776/05568 (85.78%) | loss: 2.443195 | lrm: 0.26 | dt: 1198.84ms | tok/sec: 874,656 | bf16_mfu: 53.73 | total time: 94.99m | eta: 15.8m step 04777/05568 (85.79%) | loss: 2.439845 | lrm: 0.26 | dt: 1196.94ms | tok/sec: 876,048 | bf16_mfu: 53.81 | total time: 95.01m | eta: 15.7m step 04778/05568 (85.81%) | loss: 2.442658 | lrm: 0.26 | dt: 1196.78ms | tok/sec: 876,164 | bf16_mfu: 53.82 | total time: 95.03m | eta: 15.7m step 04779/05568 (85.83%) | loss: 2.431597 | lrm: 0.26 | dt: 1193.15ms | tok/sec: 878,827 | bf16_mfu: 53.98 | total time: 95.05m | eta: 15.7m step 04780/05568 (85.85%) | loss: 2.432844 | lrm: 0.26 | dt: 1194.25ms | tok/sec: 878,020 | bf16_mfu: 53.93 | total time: 95.07m | eta: 15.7m step 04781/05568 (85.87%) | loss: 2.424999 | lrm: 0.26 | dt: 1194.34ms | tok/sec: 877,953 | bf16_mfu: 53.93 | total time: 95.09m | eta: 15.7m step 04782/05568 (85.88%) | loss: 2.420618 | lrm: 0.26 | dt: 1191.47ms | tok/sec: 880,070 | bf16_mfu: 54.06 | total time: 95.11m | eta: 15.6m step 04783/05568 (85.90%) | loss: 2.418260 | lrm: 0.26 | dt: 1191.77ms | tok/sec: 879,846 | bf16_mfu: 54.05 | total time: 95.13m | eta: 15.6m step 04784/05568 (85.92%) | loss: 2.414562 | lrm: 0.26 | dt: 1194.97ms | tok/sec: 877,490 | bf16_mfu: 53.90 | total time: 95.15m | eta: 15.6m step 04785/05568 (85.94%) | loss: 2.407586 | lrm: 0.26 | dt: 1193.16ms | tok/sec: 878,822 | bf16_mfu: 53.98 | total time: 95.17m | eta: 15.6m step 04786/05568 (85.96%) | loss: 2.391915 | lrm: 0.26 | dt: 1192.29ms | tok/sec: 879,462 | bf16_mfu: 54.02 | total time: 95.19m | eta: 15.6m step 04787/05568 (85.97%) | loss: 2.394914 | lrm: 0.26 | dt: 1194.38ms | tok/sec: 877,925 | bf16_mfu: 53.93 | total time: 95.21m | eta: 15.5m step 04788/05568 (85.99%) | loss: 2.395364 | lrm: 0.25 | dt: 1196.56ms | tok/sec: 876,323 | bf16_mfu: 53.83 | total time: 95.23m | eta: 15.5m step 04789/05568 (86.01%) | loss: 2.394471 | lrm: 0.25 | dt: 1197.24ms | tok/sec: 875,827 | bf16_mfu: 53.80 | total time: 95.25m | eta: 15.5m step 04790/05568 (86.03%) | loss: 2.396009 | lrm: 0.25 | dt: 1189.23ms | tok/sec: 881,729 | bf16_mfu: 54.16 | total time: 95.27m | eta: 15.5m step 04791/05568 (86.05%) | loss: 2.397171 | lrm: 0.25 | dt: 1192.22ms | tok/sec: 879,512 | bf16_mfu: 54.03 | total time: 95.29m | eta: 15.5m step 04792/05568 (86.06%) | loss: 2.398358 | lrm: 0.25 | dt: 1190.77ms | tok/sec: 880,587 | bf16_mfu: 54.09 | total time: 95.31m | eta: 15.4m step 04793/05568 (86.08%) | loss: 2.385135 | lrm: 0.25 | dt: 1192.48ms | tok/sec: 879,321 | bf16_mfu: 54.01 | total time: 95.33m | eta: 15.4m step 04794/05568 (86.10%) | loss: 2.394223 | lrm: 0.25 | dt: 1196.02ms | tok/sec: 876,724 | bf16_mfu: 53.86 | total time: 95.35m | eta: 15.4m step 04795/05568 (86.12%) | loss: 2.394086 | lrm: 0.25 | dt: 1191.69ms | tok/sec: 879,903 | bf16_mfu: 54.05 | total time: 95.37m | eta: 15.4m step 04796/05568 (86.14%) | loss: 2.394909 | lrm: 0.25 | dt: 1194.55ms | tok/sec: 877,800 | bf16_mfu: 53.92 | total time: 95.39m | eta: 15.4m step 04797/05568 (86.15%) | loss: 2.398120 | lrm: 0.25 | dt: 1194.30ms | tok/sec: 877,982 | bf16_mfu: 53.93 | total time: 95.41m | eta: 15.3m step 04798/05568 (86.17%) | loss: 2.407933 | lrm: 0.25 | dt: 1195.50ms | tok/sec: 877,102 | bf16_mfu: 53.88 | total time: 95.43m | eta: 15.3m step 04799/05568 (86.19%) | loss: 2.404075 | lrm: 0.25 | dt: 1193.09ms | tok/sec: 878,872 | bf16_mfu: 53.99 | total time: 95.45m | eta: 15.3m step 04800/05568 (86.21%) | loss: 2.406658 | lrm: 0.25 | dt: 1194.90ms | tok/sec: 877,544 | bf16_mfu: 53.91 | total time: 95.47m | eta: 15.3m step 04801/05568 (86.22%) | loss: 2.414800 | lrm: 0.25 | dt: 1200.76ms | tok/sec: 873,261 | bf16_mfu: 53.64 | total time: 95.49m | eta: 15.3m step 04802/05568 (86.24%) | loss: 2.407350 | lrm: 0.25 | dt: 1193.93ms | tok/sec: 878,252 | bf16_mfu: 53.95 | total time: 95.51m | eta: 15.2m step 04803/05568 (86.26%) | loss: 2.397122 | lrm: 0.25 | dt: 1194.66ms | tok/sec: 877,718 | bf16_mfu: 53.92 | total time: 95.53m | eta: 15.2m step 04804/05568 (86.28%) | loss: 2.393925 | lrm: 0.25 | dt: 1202.74ms | tok/sec: 871,819 | bf16_mfu: 53.55 | total time: 95.55m | eta: 15.2m step 04805/05568 (86.30%) | loss: 2.386434 | lrm: 0.25 | dt: 1194.49ms | tok/sec: 877,846 | bf16_mfu: 53.92 | total time: 95.57m | eta: 15.2m step 04806/05568 (86.31%) | loss: 2.384793 | lrm: 0.25 | dt: 1195.89ms | tok/sec: 876,813 | bf16_mfu: 53.86 | total time: 95.59m | eta: 15.2m step 04807/05568 (86.33%) | loss: 2.373916 | lrm: 0.25 | dt: 1196.21ms | tok/sec: 876,579 | bf16_mfu: 53.85 | total time: 95.61m | eta: 15.1m step 04808/05568 (86.35%) | loss: 2.378728 | lrm: 0.25 | dt: 1194.65ms | tok/sec: 877,725 | bf16_mfu: 53.92 | total time: 95.63m | eta: 15.1m step 04809/05568 (86.37%) | loss: 2.375832 | lrm: 0.25 | dt: 1190.56ms | tok/sec: 880,740 | bf16_mfu: 54.10 | total time: 95.65m | eta: 15.1m step 04810/05568 (86.39%) | loss: 2.383986 | lrm: 0.25 | dt: 1196.06ms | tok/sec: 876,693 | bf16_mfu: 53.85 | total time: 95.67m | eta: 15.1m step 04811/05568 (86.40%) | loss: 2.375508 | lrm: 0.25 | dt: 1196.28ms | tok/sec: 876,533 | bf16_mfu: 53.84 | total time: 95.69m | eta: 15.1m step 04812/05568 (86.42%) | loss: 2.383674 | lrm: 0.25 | dt: 1194.52ms | tok/sec: 877,822 | bf16_mfu: 53.92 | total time: 95.71m | eta: 15.0m step 04813/05568 (86.44%) | loss: 2.382852 | lrm: 0.25 | dt: 1190.79ms | tok/sec: 880,573 | bf16_mfu: 54.09 | total time: 95.73m | eta: 15.0m step 04814/05568 (86.46%) | loss: 2.379840 | lrm: 0.25 | dt: 1198.79ms | tok/sec: 874,692 | bf16_mfu: 53.73 | total time: 95.75m | eta: 15.0m step 04815/05568 (86.48%) | loss: 2.389208 | lrm: 0.25 | dt: 1202.20ms | tok/sec: 872,215 | bf16_mfu: 53.58 | total time: 95.77m | eta: 15.0m step 04816/05568 (86.49%) | loss: 2.392659 | lrm: 0.25 | dt: 1194.10ms | tok/sec: 878,128 | bf16_mfu: 53.94 | total time: 95.79m | eta: 15.0m step 04817/05568 (86.51%) | loss: 2.394624 | lrm: 0.25 | dt: 1196.20ms | tok/sec: 876,586 | bf16_mfu: 53.85 | total time: 95.81m | eta: 14.9m step 04818/05568 (86.53%) | loss: 2.402610 | lrm: 0.25 | dt: 1196.85ms | tok/sec: 876,114 | bf16_mfu: 53.82 | total time: 95.83m | eta: 14.9m step 04819/05568 (86.55%) | loss: 2.406519 | lrm: 0.25 | dt: 1193.97ms | tok/sec: 878,223 | bf16_mfu: 53.95 | total time: 95.85m | eta: 14.9m step 04820/05568 (86.57%) | loss: 2.405724 | lrm: 0.25 | dt: 1199.21ms | tok/sec: 874,392 | bf16_mfu: 53.71 | total time: 95.87m | eta: 14.9m step 04821/05568 (86.58%) | loss: 2.399719 | lrm: 0.25 | dt: 1192.58ms | tok/sec: 879,251 | bf16_mfu: 54.01 | total time: 95.89m | eta: 14.9m step 04822/05568 (86.60%) | loss: 2.404538 | lrm: 0.25 | dt: 1193.89ms | tok/sec: 878,283 | bf16_mfu: 53.95 | total time: 95.91m | eta: 14.8m step 04823/05568 (86.62%) | loss: 2.409289 | lrm: 0.25 | dt: 1196.64ms | tok/sec: 876,263 | bf16_mfu: 53.83 | total time: 95.93m | eta: 14.8m step 04824/05568 (86.64%) | loss: 2.411435 | lrm: 0.25 | dt: 1193.81ms | tok/sec: 878,343 | bf16_mfu: 53.95 | total time: 95.95m | eta: 14.8m step 04825/05568 (86.66%) | loss: 2.408986 | lrm: 0.25 | dt: 1198.15ms | tok/sec: 875,165 | bf16_mfu: 53.76 | total time: 95.97m | eta: 14.8m step 04826/05568 (86.67%) | loss: 2.398875 | lrm: 0.24 | dt: 1195.46ms | tok/sec: 877,132 | bf16_mfu: 53.88 | total time: 95.99m | eta: 14.8m step 04827/05568 (86.69%) | loss: 2.402270 | lrm: 0.24 | dt: 1192.11ms | tok/sec: 879,596 | bf16_mfu: 54.03 | total time: 96.01m | eta: 14.7m step 04828/05568 (86.71%) | loss: 2.395846 | lrm: 0.24 | dt: 1195.81ms | tok/sec: 876,872 | bf16_mfu: 53.86 | total time: 96.03m | eta: 14.7m step 04829/05568 (86.73%) | loss: 2.389649 | lrm: 0.24 | dt: 1192.99ms | tok/sec: 878,945 | bf16_mfu: 53.99 | total time: 96.05m | eta: 14.7m step 04830/05568 (86.75%) | loss: 2.387214 | lrm: 0.24 | dt: 1195.88ms | tok/sec: 876,825 | bf16_mfu: 53.86 | total time: 96.07m | eta: 14.7m step 04831/05568 (86.76%) | loss: 2.377831 | lrm: 0.24 | dt: 1189.28ms | tok/sec: 881,689 | bf16_mfu: 54.16 | total time: 96.09m | eta: 14.7m step 04832/05568 (86.78%) | loss: 2.374884 | lrm: 0.24 | dt: 1191.64ms | tok/sec: 879,941 | bf16_mfu: 54.05 | total time: 96.11m | eta: 14.6m step 04833/05568 (86.80%) | loss: 2.377102 | lrm: 0.24 | dt: 1194.08ms | tok/sec: 878,142 | bf16_mfu: 53.94 | total time: 96.13m | eta: 14.6m step 04834/05568 (86.82%) | loss: 2.374659 | lrm: 0.24 | dt: 1196.04ms | tok/sec: 876,705 | bf16_mfu: 53.85 | total time: 96.15m | eta: 14.6m step 04835/05568 (86.84%) | loss: 2.377839 | lrm: 0.24 | dt: 1192.65ms | tok/sec: 879,200 | bf16_mfu: 54.01 | total time: 96.17m | eta: 14.6m step 04836/05568 (86.85%) | loss: 2.396825 | lrm: 0.24 | dt: 1197.04ms | tok/sec: 875,976 | bf16_mfu: 53.81 | total time: 96.19m | eta: 14.6m step 04837/05568 (86.87%) | loss: 2.392029 | lrm: 0.24 | dt: 1192.87ms | tok/sec: 879,033 | bf16_mfu: 54.00 | total time: 96.21m | eta: 14.5m step 04838/05568 (86.89%) | loss: 2.383035 | lrm: 0.24 | dt: 1198.52ms | tok/sec: 874,890 | bf16_mfu: 53.74 | total time: 96.23m | eta: 14.5m step 04839/05568 (86.91%) | loss: 2.384348 | lrm: 0.24 | dt: 1193.20ms | tok/sec: 878,794 | bf16_mfu: 53.98 | total time: 96.25m | eta: 14.5m step 04840/05568 (86.93%) | loss: 2.388676 | lrm: 0.24 | dt: 1195.69ms | tok/sec: 876,961 | bf16_mfu: 53.87 | total time: 96.26m | eta: 14.5m step 04841/05568 (86.94%) | loss: 2.387001 | lrm: 0.24 | dt: 1194.82ms | tok/sec: 877,602 | bf16_mfu: 53.91 | total time: 96.28m | eta: 14.5m step 04842/05568 (86.96%) | loss: 2.379347 | lrm: 0.24 | dt: 1193.91ms | tok/sec: 878,268 | bf16_mfu: 53.95 | total time: 96.30m | eta: 14.4m step 04843/05568 (86.98%) | loss: 2.377705 | lrm: 0.24 | dt: 1198.99ms | tok/sec: 874,550 | bf16_mfu: 53.72 | total time: 96.32m | eta: 14.4m step 04844/05568 (87.00%) | loss: 2.379353 | lrm: 0.24 | dt: 1195.80ms | tok/sec: 876,880 | bf16_mfu: 53.86 | total time: 96.34m | eta: 14.4m step 04845/05568 (87.02%) | loss: 2.370317 | lrm: 0.24 | dt: 1193.09ms | tok/sec: 878,874 | bf16_mfu: 53.99 | total time: 96.36m | eta: 14.4m step 04846/05568 (87.03%) | loss: 2.388399 | lrm: 0.24 | dt: 1195.24ms | tok/sec: 877,294 | bf16_mfu: 53.89 | total time: 96.38m | eta: 14.4m step 04847/05568 (87.05%) | loss: 2.383305 | lrm: 0.24 | dt: 1201.61ms | tok/sec: 872,639 | bf16_mfu: 53.60 | total time: 96.40m | eta: 14.4m step 04848/05568 (87.07%) | loss: 2.394296 | lrm: 0.24 | dt: 1190.89ms | tok/sec: 880,494 | bf16_mfu: 54.09 | total time: 96.42m | eta: 14.3m step 04849/05568 (87.09%) | loss: 2.391815 | lrm: 0.24 | dt: 1195.19ms | tok/sec: 877,330 | bf16_mfu: 53.89 | total time: 96.44m | eta: 14.3m step 04850/05568 (87.10%) | loss: 2.394859 | lrm: 0.24 | dt: 1195.80ms | tok/sec: 876,884 | bf16_mfu: 53.86 | total time: 96.46m | eta: 14.3m step 04851/05568 (87.12%) | loss: 2.390694 | lrm: 0.24 | dt: 1196.92ms | tok/sec: 876,064 | bf16_mfu: 53.81 | total time: 96.48m | eta: 14.3m step 04852/05568 (87.14%) | loss: 2.386314 | lrm: 0.24 | dt: 1190.63ms | tok/sec: 880,692 | bf16_mfu: 54.10 | total time: 96.50m | eta: 14.3m step 04853/05568 (87.16%) | loss: 2.388489 | lrm: 0.24 | dt: 1196.03ms | tok/sec: 876,715 | bf16_mfu: 53.85 | total time: 96.52m | eta: 14.2m step 04854/05568 (87.18%) | loss: 2.391627 | lrm: 0.24 | dt: 1193.71ms | tok/sec: 878,415 | bf16_mfu: 53.96 | total time: 96.54m | eta: 14.2m step 04855/05568 (87.19%) | loss: 2.407349 | lrm: 0.24 | dt: 1191.47ms | tok/sec: 880,065 | bf16_mfu: 54.06 | total time: 96.56m | eta: 14.2m step 04856/05568 (87.21%) | loss: 2.410910 | lrm: 0.24 | dt: 1190.44ms | tok/sec: 880,827 | bf16_mfu: 54.11 | total time: 96.58m | eta: 14.2m step 04857/05568 (87.23%) | loss: 2.420351 | lrm: 0.24 | dt: 1200.00ms | tok/sec: 873,812 | bf16_mfu: 53.68 | total time: 96.60m | eta: 14.2m step 04858/05568 (87.25%) | loss: 2.413510 | lrm: 0.24 | dt: 1191.04ms | tok/sec: 880,385 | bf16_mfu: 54.08 | total time: 96.62m | eta: 14.1m step 04859/05568 (87.27%) | loss: 2.400387 | lrm: 0.24 | dt: 1197.57ms | tok/sec: 875,589 | bf16_mfu: 53.79 | total time: 96.64m | eta: 14.1m step 04860/05568 (87.28%) | loss: 2.398454 | lrm: 0.24 | dt: 1194.84ms | tok/sec: 877,590 | bf16_mfu: 53.91 | total time: 96.66m | eta: 14.1m step 04861/05568 (87.30%) | loss: 2.397951 | lrm: 0.24 | dt: 1196.54ms | tok/sec: 876,338 | bf16_mfu: 53.83 | total time: 96.68m | eta: 14.1m step 04862/05568 (87.32%) | loss: 2.398053 | lrm: 0.24 | dt: 1191.74ms | tok/sec: 879,872 | bf16_mfu: 54.05 | total time: 96.70m | eta: 14.1m step 04863/05568 (87.34%) | loss: 2.398267 | lrm: 0.24 | dt: 1198.70ms | tok/sec: 874,763 | bf16_mfu: 53.73 | total time: 96.72m | eta: 14.0m step 04864/05568 (87.36%) | loss: 2.401198 | lrm: 0.23 | dt: 1192.85ms | tok/sec: 879,051 | bf16_mfu: 54.00 | total time: 96.74m | eta: 14.0m step 04865/05568 (87.37%) | loss: 2.407884 | lrm: 0.23 | dt: 1196.02ms | tok/sec: 876,723 | bf16_mfu: 53.86 | total time: 96.76m | eta: 14.0m step 04866/05568 (87.39%) | loss: 2.396558 | lrm: 0.23 | dt: 1199.58ms | tok/sec: 874,118 | bf16_mfu: 53.70 | total time: 96.78m | eta: 14.0m step 04867/05568 (87.41%) | loss: 2.395392 | lrm: 0.23 | dt: 1199.54ms | tok/sec: 874,144 | bf16_mfu: 53.70 | total time: 96.80m | eta: 14.0m step 04868/05568 (87.43%) | loss: 2.403717 | lrm: 0.23 | dt: 1194.82ms | tok/sec: 877,600 | bf16_mfu: 53.91 | total time: 96.82m | eta: 13.9m step 04869/05568 (87.45%) | loss: 2.404410 | lrm: 0.23 | dt: 1195.86ms | tok/sec: 876,840 | bf16_mfu: 53.86 | total time: 96.84m | eta: 13.9m step 04870/05568 (87.46%) | loss: 2.397356 | lrm: 0.23 | dt: 1192.81ms | tok/sec: 879,083 | bf16_mfu: 54.00 | total time: 96.86m | eta: 13.9m step 04871/05568 (87.48%) | loss: 2.394181 | lrm: 0.23 | dt: 1199.59ms | tok/sec: 874,108 | bf16_mfu: 53.69 | total time: 96.88m | eta: 13.9m step 04872/05568 (87.50%) | loss: 2.390338 | lrm: 0.23 | dt: 1194.39ms | tok/sec: 877,915 | bf16_mfu: 53.93 | total time: 96.90m | eta: 13.9m step 04873/05568 (87.52%) | loss: 2.392941 | lrm: 0.23 | dt: 1196.21ms | tok/sec: 876,584 | bf16_mfu: 53.85 | total time: 96.92m | eta: 13.8m step 04874/05568 (87.54%) | loss: 2.395197 | lrm: 0.23 | dt: 1189.00ms | tok/sec: 881,897 | bf16_mfu: 54.17 | total time: 96.94m | eta: 13.8m step 04875/05568 (87.55%) | loss: 2.394902 | lrm: 0.23 | dt: 1199.18ms | tok/sec: 874,411 | bf16_mfu: 53.71 | total time: 96.96m | eta: 13.8m step 04876/05568 (87.57%) | loss: 2.401186 | lrm: 0.23 | dt: 1197.40ms | tok/sec: 875,712 | bf16_mfu: 53.79 | total time: 96.98m | eta: 13.8m step 04877/05568 (87.59%) | loss: 2.405241 | lrm: 0.23 | dt: 1196.04ms | tok/sec: 876,709 | bf16_mfu: 53.85 | total time: 97.00m | eta: 13.8m step 04878/05568 (87.61%) | loss: 2.404210 | lrm: 0.23 | dt: 1194.74ms | tok/sec: 877,661 | bf16_mfu: 53.91 | total time: 97.02m | eta: 13.7m step 04879/05568 (87.63%) | loss: 2.410009 | lrm: 0.23 | dt: 1194.38ms | tok/sec: 877,926 | bf16_mfu: 53.93 | total time: 97.04m | eta: 13.7m step 04880/05568 (87.64%) | loss: 2.414619 | lrm: 0.23 | dt: 1194.81ms | tok/sec: 877,606 | bf16_mfu: 53.91 | total time: 97.06m | eta: 13.7m step 04881/05568 (87.66%) | loss: 2.407500 | lrm: 0.23 | dt: 1196.20ms | tok/sec: 876,589 | bf16_mfu: 53.85 | total time: 97.08m | eta: 13.7m step 04882/05568 (87.68%) | loss: 2.403632 | lrm: 0.23 | dt: 1193.74ms | tok/sec: 878,392 | bf16_mfu: 53.96 | total time: 97.10m | eta: 13.7m step 04883/05568 (87.70%) | loss: 2.405613 | lrm: 0.23 | dt: 1194.01ms | tok/sec: 878,199 | bf16_mfu: 53.95 | total time: 97.12m | eta: 13.6m step 04884/05568 (87.72%) | loss: 2.399700 | lrm: 0.23 | dt: 1196.07ms | tok/sec: 876,682 | bf16_mfu: 53.85 | total time: 97.14m | eta: 13.6m step 04885/05568 (87.73%) | loss: 2.384893 | lrm: 0.23 | dt: 1198.46ms | tok/sec: 874,935 | bf16_mfu: 53.75 | total time: 97.16m | eta: 13.6m step 04886/05568 (87.75%) | loss: 2.388411 | lrm: 0.23 | dt: 1196.57ms | tok/sec: 876,318 | bf16_mfu: 53.83 | total time: 97.18m | eta: 13.6m step 04887/05568 (87.77%) | loss: 2.392068 | lrm: 0.23 | dt: 1198.08ms | tok/sec: 875,216 | bf16_mfu: 53.76 | total time: 97.20m | eta: 13.6m step 04888/05568 (87.79%) | loss: 2.387313 | lrm: 0.23 | dt: 1196.56ms | tok/sec: 876,323 | bf16_mfu: 53.83 | total time: 97.22m | eta: 13.5m step 04889/05568 (87.81%) | loss: 2.385143 | lrm: 0.23 | dt: 1196.20ms | tok/sec: 876,589 | bf16_mfu: 53.85 | total time: 97.24m | eta: 13.5m step 04890/05568 (87.82%) | loss: 2.393831 | lrm: 0.23 | dt: 1197.04ms | tok/sec: 875,970 | bf16_mfu: 53.81 | total time: 97.26m | eta: 13.5m step 04891/05568 (87.84%) | loss: 2.392668 | lrm: 0.23 | dt: 1195.93ms | tok/sec: 876,788 | bf16_mfu: 53.86 | total time: 97.28m | eta: 13.5m step 04892/05568 (87.86%) | loss: 2.405136 | lrm: 0.23 | dt: 1196.40ms | tok/sec: 876,443 | bf16_mfu: 53.84 | total time: 97.30m | eta: 13.5m step 04893/05568 (87.88%) | loss: 2.409043 | lrm: 0.23 | dt: 1197.06ms | tok/sec: 875,961 | bf16_mfu: 53.81 | total time: 97.32m | eta: 13.4m step 04894/05568 (87.90%) | loss: 2.404607 | lrm: 0.23 | dt: 1195.51ms | tok/sec: 877,095 | bf16_mfu: 53.88 | total time: 97.34m | eta: 13.4m step 04895/05568 (87.91%) | loss: 2.396794 | lrm: 0.23 | dt: 1194.15ms | tok/sec: 878,092 | bf16_mfu: 53.94 | total time: 97.36m | eta: 13.4m step 04896/05568 (87.93%) | loss: 2.389187 | lrm: 0.23 | dt: 1197.20ms | tok/sec: 875,854 | bf16_mfu: 53.80 | total time: 97.38m | eta: 13.4m step 04897/05568 (87.95%) | loss: 2.374634 | lrm: 0.23 | dt: 1189.89ms | tok/sec: 881,235 | bf16_mfu: 54.13 | total time: 97.40m | eta: 13.4m step 04898/05568 (87.97%) | loss: 2.372261 | lrm: 0.23 | dt: 1197.53ms | tok/sec: 875,618 | bf16_mfu: 53.79 | total time: 97.42m | eta: 13.3m step 04899/05568 (87.98%) | loss: 2.382527 | lrm: 0.23 | dt: 1192.17ms | tok/sec: 879,549 | bf16_mfu: 54.03 | total time: 97.44m | eta: 13.3m step 04900/05568 (88.00%) | loss: 2.379619 | lrm: 0.23 | dt: 1192.51ms | tok/sec: 879,300 | bf16_mfu: 54.01 | total time: 97.46m | eta: 13.3m step 04901/05568 (88.02%) | loss: 2.389688 | lrm: 0.23 | dt: 1194.95ms | tok/sec: 877,509 | bf16_mfu: 53.90 | total time: 97.48m | eta: 13.3m step 04902/05568 (88.04%) | loss: 2.401101 | lrm: 0.22 | dt: 1197.77ms | tok/sec: 875,437 | bf16_mfu: 53.78 | total time: 97.50m | eta: 13.3m step 04903/05568 (88.06%) | loss: 2.409125 | lrm: 0.22 | dt: 1194.53ms | tok/sec: 877,814 | bf16_mfu: 53.92 | total time: 97.52m | eta: 13.2m step 04904/05568 (88.07%) | loss: 2.408245 | lrm: 0.22 | dt: 1195.29ms | tok/sec: 877,259 | bf16_mfu: 53.89 | total time: 97.54m | eta: 13.2m step 04905/05568 (88.09%) | loss: 2.403391 | lrm: 0.22 | dt: 1195.85ms | tok/sec: 876,842 | bf16_mfu: 53.86 | total time: 97.56m | eta: 13.2m step 04906/05568 (88.11%) | loss: 2.406417 | lrm: 0.22 | dt: 1192.21ms | tok/sec: 879,522 | bf16_mfu: 54.03 | total time: 97.58m | eta: 13.2m step 04907/05568 (88.13%) | loss: 2.406372 | lrm: 0.22 | dt: 1198.17ms | tok/sec: 875,147 | bf16_mfu: 53.76 | total time: 97.60m | eta: 13.2m step 04908/05568 (88.15%) | loss: 2.403047 | lrm: 0.22 | dt: 1197.85ms | tok/sec: 875,378 | bf16_mfu: 53.77 | total time: 97.62m | eta: 13.1m step 04909/05568 (88.16%) | loss: 2.401002 | lrm: 0.22 | dt: 1193.81ms | tok/sec: 878,341 | bf16_mfu: 53.95 | total time: 97.64m | eta: 13.1m step 04910/05568 (88.18%) | loss: 2.403047 | lrm: 0.22 | dt: 1188.92ms | tok/sec: 881,956 | bf16_mfu: 54.18 | total time: 97.66m | eta: 13.1m step 04911/05568 (88.20%) | loss: 2.400411 | lrm: 0.22 | dt: 1199.14ms | tok/sec: 874,443 | bf16_mfu: 53.72 | total time: 97.68m | eta: 13.1m step 04912/05568 (88.22%) | loss: 2.404733 | lrm: 0.22 | dt: 1199.12ms | tok/sec: 874,451 | bf16_mfu: 53.72 | total time: 97.70m | eta: 13.1m step 04913/05568 (88.24%) | loss: 2.409138 | lrm: 0.22 | dt: 1196.00ms | tok/sec: 876,734 | bf16_mfu: 53.86 | total time: 97.72m | eta: 13.0m step 04914/05568 (88.25%) | loss: 2.408963 | lrm: 0.22 | dt: 1195.52ms | tok/sec: 877,091 | bf16_mfu: 53.88 | total time: 97.74m | eta: 13.0m step 04915/05568 (88.27%) | loss: 2.398082 | lrm: 0.22 | dt: 1193.82ms | tok/sec: 878,335 | bf16_mfu: 53.95 | total time: 97.76m | eta: 13.0m step 04916/05568 (88.29%) | loss: 2.393262 | lrm: 0.22 | dt: 1195.65ms | tok/sec: 876,995 | bf16_mfu: 53.87 | total time: 97.78m | eta: 13.0m step 04917/05568 (88.31%) | loss: 2.390708 | lrm: 0.22 | dt: 1196.60ms | tok/sec: 876,293 | bf16_mfu: 53.83 | total time: 97.80m | eta: 13.0m step 04918/05568 (88.33%) | loss: 2.385289 | lrm: 0.22 | dt: 1195.21ms | tok/sec: 877,317 | bf16_mfu: 53.89 | total time: 97.82m | eta: 12.9m step 04919/05568 (88.34%) | loss: 2.379486 | lrm: 0.22 | dt: 1193.52ms | tok/sec: 878,559 | bf16_mfu: 53.97 | total time: 97.84m | eta: 12.9m step 04920/05568 (88.36%) | loss: 2.379834 | lrm: 0.22 | dt: 1192.77ms | tok/sec: 879,112 | bf16_mfu: 54.00 | total time: 97.86m | eta: 12.9m step 04921/05568 (88.38%) | loss: 2.385494 | lrm: 0.22 | dt: 1193.86ms | tok/sec: 878,310 | bf16_mfu: 53.95 | total time: 97.88m | eta: 12.9m step 04922/05568 (88.40%) | loss: 2.391950 | lrm: 0.22 | dt: 1194.77ms | tok/sec: 877,640 | bf16_mfu: 53.91 | total time: 97.90m | eta: 12.9m step 04923/05568 (88.42%) | loss: 2.389801 | lrm: 0.22 | dt: 1198.96ms | tok/sec: 874,574 | bf16_mfu: 53.72 | total time: 97.92m | eta: 12.8m step 04924/05568 (88.43%) | loss: 2.384711 | lrm: 0.22 | dt: 1194.59ms | tok/sec: 877,771 | bf16_mfu: 53.92 | total time: 97.94m | eta: 12.8m step 04925/05568 (88.45%) | loss: 2.382897 | lrm: 0.22 | dt: 1196.28ms | tok/sec: 876,533 | bf16_mfu: 53.84 | total time: 97.96m | eta: 12.8m step 04926/05568 (88.47%) | loss: 2.395502 | lrm: 0.22 | dt: 1199.57ms | tok/sec: 874,123 | bf16_mfu: 53.70 | total time: 97.98m | eta: 12.8m step 04927/05568 (88.49%) | loss: 2.393275 | lrm: 0.22 | dt: 1196.87ms | tok/sec: 876,098 | bf16_mfu: 53.82 | total time: 98.00m | eta: 12.8m step 04928/05568 (88.51%) | loss: 2.393419 | lrm: 0.22 | dt: 1194.25ms | tok/sec: 878,018 | bf16_mfu: 53.93 | total time: 98.02m | eta: 12.7m step 04929/05568 (88.52%) | loss: 2.391208 | lrm: 0.22 | dt: 1193.50ms | tok/sec: 878,568 | bf16_mfu: 53.97 | total time: 98.04m | eta: 12.7m step 04930/05568 (88.54%) | loss: 2.386279 | lrm: 0.22 | dt: 1197.03ms | tok/sec: 875,978 | bf16_mfu: 53.81 | total time: 98.06m | eta: 12.7m step 04931/05568 (88.56%) | loss: 2.393842 | lrm: 0.22 | dt: 1194.78ms | tok/sec: 877,631 | bf16_mfu: 53.91 | total time: 98.08m | eta: 12.7m step 04932/05568 (88.58%) | loss: 2.399118 | lrm: 0.22 | dt: 1196.85ms | tok/sec: 876,112 | bf16_mfu: 53.82 | total time: 98.10m | eta: 12.7m step 04933/05568 (88.60%) | loss: 2.403760 | lrm: 0.22 | dt: 1194.04ms | tok/sec: 878,178 | bf16_mfu: 53.94 | total time: 98.12m | eta: 12.6m step 04934/05568 (88.61%) | loss: 2.397965 | lrm: 0.22 | dt: 1192.80ms | tok/sec: 879,085 | bf16_mfu: 54.00 | total time: 98.14m | eta: 12.6m step 04935/05568 (88.63%) | loss: 2.401917 | lrm: 0.22 | dt: 1196.48ms | tok/sec: 876,386 | bf16_mfu: 53.83 | total time: 98.16m | eta: 12.6m step 04936/05568 (88.65%) | loss: 2.399227 | lrm: 0.22 | dt: 1199.09ms | tok/sec: 874,474 | bf16_mfu: 53.72 | total time: 98.18m | eta: 12.6m step 04937/05568 (88.67%) | loss: 2.419449 | lrm: 0.22 | dt: 1195.98ms | tok/sec: 876,752 | bf16_mfu: 53.86 | total time: 98.20m | eta: 12.6m step 04938/05568 (88.69%) | loss: 2.423515 | lrm: 0.22 | dt: 1195.11ms | tok/sec: 877,389 | bf16_mfu: 53.90 | total time: 98.22m | eta: 12.5m step 04939/05568 (88.70%) | loss: 2.431088 | lrm: 0.22 | dt: 1193.83ms | tok/sec: 878,329 | bf16_mfu: 53.95 | total time: 98.24m | eta: 12.5m step 04940/05568 (88.72%) | loss: 2.425240 | lrm: 0.21 | dt: 1193.68ms | tok/sec: 878,442 | bf16_mfu: 53.96 | total time: 98.26m | eta: 12.5m step 04941/05568 (88.74%) | loss: 2.416710 | lrm: 0.21 | dt: 1199.95ms | tok/sec: 873,851 | bf16_mfu: 53.68 | total time: 98.28m | eta: 12.5m step 04942/05568 (88.76%) | loss: 2.406359 | lrm: 0.21 | dt: 1193.50ms | tok/sec: 878,572 | bf16_mfu: 53.97 | total time: 98.30m | eta: 12.5m step 04943/05568 (88.78%) | loss: 2.403514 | lrm: 0.21 | dt: 1197.45ms | tok/sec: 875,674 | bf16_mfu: 53.79 | total time: 98.32m | eta: 12.4m step 04944/05568 (88.79%) | loss: 2.401084 | lrm: 0.21 | dt: 1194.55ms | tok/sec: 877,800 | bf16_mfu: 53.92 | total time: 98.34m | eta: 12.4m step 04945/05568 (88.81%) | loss: 2.409357 | lrm: 0.21 | dt: 1193.09ms | tok/sec: 878,876 | bf16_mfu: 53.99 | total time: 98.36m | eta: 12.4m step 04946/05568 (88.83%) | loss: 2.420122 | lrm: 0.21 | dt: 1196.25ms | tok/sec: 876,553 | bf16_mfu: 53.84 | total time: 98.38m | eta: 12.4m step 04947/05568 (88.85%) | loss: 2.426456 | lrm: 0.21 | dt: 1188.96ms | tok/sec: 881,928 | bf16_mfu: 54.17 | total time: 98.40m | eta: 12.4m step 04948/05568 (88.86%) | loss: 2.417544 | lrm: 0.21 | dt: 1194.45ms | tok/sec: 877,872 | bf16_mfu: 53.93 | total time: 98.42m | eta: 12.3m step 04949/05568 (88.88%) | loss: 2.409477 | lrm: 0.21 | dt: 1196.50ms | tok/sec: 876,366 | bf16_mfu: 53.83 | total time: 98.44m | eta: 12.3m step 04950/05568 (88.90%) | loss: 2.402195 | lrm: 0.21 | dt: 1195.88ms | tok/sec: 876,823 | bf16_mfu: 53.86 | total time: 98.46m | eta: 12.3m step 04951/05568 (88.92%) | loss: 2.405715 | lrm: 0.21 | dt: 1191.90ms | tok/sec: 879,748 | bf16_mfu: 54.04 | total time: 98.48m | eta: 12.3m step 04952/05568 (88.94%) | loss: 2.412305 | lrm: 0.21 | dt: 1191.54ms | tok/sec: 880,015 | bf16_mfu: 54.06 | total time: 98.50m | eta: 12.3m step 04953/05568 (88.95%) | loss: 2.401445 | lrm: 0.21 | dt: 1190.91ms | tok/sec: 880,482 | bf16_mfu: 54.09 | total time: 98.52m | eta: 12.2m step 04954/05568 (88.97%) | loss: 2.397680 | lrm: 0.21 | dt: 1197.41ms | tok/sec: 875,703 | bf16_mfu: 53.79 | total time: 98.54m | eta: 12.2m step 04955/05568 (88.99%) | loss: 2.408182 | lrm: 0.21 | dt: 1194.14ms | tok/sec: 878,102 | bf16_mfu: 53.94 | total time: 98.56m | eta: 12.2m step 04956/05568 (89.01%) | loss: 2.403907 | lrm: 0.21 | dt: 1197.05ms | tok/sec: 875,967 | bf16_mfu: 53.81 | total time: 98.58m | eta: 12.2m step 04957/05568 (89.03%) | loss: 2.407160 | lrm: 0.21 | dt: 1194.78ms | tok/sec: 877,631 | bf16_mfu: 53.91 | total time: 98.60m | eta: 12.2m step 04958/05568 (89.04%) | loss: 2.403990 | lrm: 0.21 | dt: 1196.68ms | tok/sec: 876,234 | bf16_mfu: 53.83 | total time: 98.62m | eta: 12.1m step 04959/05568 (89.06%) | loss: 2.401344 | lrm: 0.21 | dt: 1197.88ms | tok/sec: 875,361 | bf16_mfu: 53.77 | total time: 98.64m | eta: 12.1m step 04960/05568 (89.08%) | loss: 2.403922 | lrm: 0.21 | dt: 1192.73ms | tok/sec: 879,137 | bf16_mfu: 54.00 | total time: 98.66m | eta: 12.1m step 04961/05568 (89.10%) | loss: 2.415541 | lrm: 0.21 | dt: 1192.16ms | tok/sec: 879,562 | bf16_mfu: 54.03 | total time: 98.68m | eta: 12.1m step 04962/05568 (89.12%) | loss: 2.403905 | lrm: 0.21 | dt: 1194.93ms | tok/sec: 877,519 | bf16_mfu: 53.90 | total time: 98.70m | eta: 12.1m step 04963/05568 (89.13%) | loss: 2.404258 | lrm: 0.21 | dt: 1196.99ms | tok/sec: 876,007 | bf16_mfu: 53.81 | total time: 98.72m | eta: 12.0m step 04964/05568 (89.15%) | loss: 2.395944 | lrm: 0.21 | dt: 1195.50ms | tok/sec: 877,105 | bf16_mfu: 53.88 | total time: 98.74m | eta: 12.0m step 04965/05568 (89.17%) | loss: 2.398756 | lrm: 0.21 | dt: 1192.70ms | tok/sec: 879,162 | bf16_mfu: 54.00 | total time: 98.76m | eta: 12.0m step 04966/05568 (89.19%) | loss: 2.400721 | lrm: 0.21 | dt: 1194.78ms | tok/sec: 877,627 | bf16_mfu: 53.91 | total time: 98.78m | eta: 12.0m step 04967/05568 (89.21%) | loss: 2.395587 | lrm: 0.21 | dt: 1196.21ms | tok/sec: 876,583 | bf16_mfu: 53.85 | total time: 98.79m | eta: 12.0m step 04968/05568 (89.22%) | loss: 2.389058 | lrm: 0.21 | dt: 1198.87ms | tok/sec: 874,636 | bf16_mfu: 53.73 | total time: 98.81m | eta: 11.9m step 04969/05568 (89.24%) | loss: 2.375811 | lrm: 0.21 | dt: 1196.33ms | tok/sec: 876,494 | bf16_mfu: 53.84 | total time: 98.83m | eta: 11.9m step 04970/05568 (89.26%) | loss: 2.372122 | lrm: 0.21 | dt: 1196.83ms | tok/sec: 876,129 | bf16_mfu: 53.82 | total time: 98.85m | eta: 11.9m step 04971/05568 (89.28%) | loss: 2.371405 | lrm: 0.21 | dt: 1194.61ms | tok/sec: 877,758 | bf16_mfu: 53.92 | total time: 98.87m | eta: 11.9m step 04972/05568 (89.30%) | loss: 2.372979 | lrm: 0.21 | dt: 1192.70ms | tok/sec: 879,164 | bf16_mfu: 54.01 | total time: 98.89m | eta: 11.9m step 04973/05568 (89.31%) | loss: 2.367506 | lrm: 0.21 | dt: 1195.74ms | tok/sec: 876,929 | bf16_mfu: 53.87 | total time: 98.91m | eta: 11.8m step 04974/05568 (89.33%) | loss: 2.357476 | lrm: 0.21 | dt: 1197.40ms | tok/sec: 875,707 | bf16_mfu: 53.79 | total time: 98.93m | eta: 11.8m step 04975/05568 (89.35%) | loss: 2.371291 | lrm: 0.21 | dt: 1193.55ms | tok/sec: 878,535 | bf16_mfu: 53.97 | total time: 98.95m | eta: 11.8m step 04976/05568 (89.37%) | loss: 2.368635 | lrm: 0.21 | dt: 1191.45ms | tok/sec: 880,084 | bf16_mfu: 54.06 | total time: 98.97m | eta: 11.8m step 04977/05568 (89.39%) | loss: 2.367786 | lrm: 0.21 | dt: 1196.47ms | tok/sec: 876,390 | bf16_mfu: 53.83 | total time: 98.99m | eta: 11.8m step 04978/05568 (89.40%) | loss: 2.378301 | lrm: 0.20 | dt: 1197.31ms | tok/sec: 875,774 | bf16_mfu: 53.80 | total time: 99.01m | eta: 11.7m step 04979/05568 (89.42%) | loss: 2.390784 | lrm: 0.20 | dt: 1191.38ms | tok/sec: 880,138 | bf16_mfu: 54.06 | total time: 99.03m | eta: 11.7m step 04980/05568 (89.44%) | loss: 2.372784 | lrm: 0.20 | dt: 1193.02ms | tok/sec: 878,922 | bf16_mfu: 53.99 | total time: 99.05m | eta: 11.7m step 04981/05568 (89.46%) | loss: 2.382943 | lrm: 0.20 | dt: 1197.39ms | tok/sec: 875,715 | bf16_mfu: 53.79 | total time: 99.07m | eta: 11.7m step 04982/05568 (89.48%) | loss: 2.384071 | lrm: 0.20 | dt: 1198.08ms | tok/sec: 875,212 | bf16_mfu: 53.76 | total time: 99.09m | eta: 11.7m step 04983/05568 (89.49%) | loss: 2.381444 | lrm: 0.20 | dt: 1189.22ms | tok/sec: 881,731 | bf16_mfu: 54.16 | total time: 99.11m | eta: 11.6m step 04984/05568 (89.51%) | loss: 2.378556 | lrm: 0.20 | dt: 1193.64ms | tok/sec: 878,468 | bf16_mfu: 53.96 | total time: 99.13m | eta: 11.6m step 04985/05568 (89.53%) | loss: 2.385139 | lrm: 0.20 | dt: 1194.49ms | tok/sec: 877,846 | bf16_mfu: 53.92 | total time: 99.15m | eta: 11.6m step 04986/05568 (89.55%) | loss: 2.381165 | lrm: 0.20 | dt: 1193.99ms | tok/sec: 878,210 | bf16_mfu: 53.95 | total time: 99.17m | eta: 11.6m step 04987/05568 (89.57%) | loss: 2.378284 | lrm: 0.20 | dt: 1193.16ms | tok/sec: 878,818 | bf16_mfu: 53.98 | total time: 99.19m | eta: 11.6m step 04988/05568 (89.58%) | loss: 2.381021 | lrm: 0.20 | dt: 1196.11ms | tok/sec: 876,658 | bf16_mfu: 53.85 | total time: 99.21m | eta: 11.5m step 04989/05568 (89.60%) | loss: 2.401765 | lrm: 0.20 | dt: 1200.05ms | tok/sec: 873,780 | bf16_mfu: 53.67 | total time: 99.23m | eta: 11.5m step 04990/05568 (89.62%) | loss: 2.402819 | lrm: 0.20 | dt: 1198.94ms | tok/sec: 874,583 | bf16_mfu: 53.72 | total time: 99.25m | eta: 11.5m step 04991/05568 (89.64%) | loss: 2.409708 | lrm: 0.20 | dt: 1196.49ms | tok/sec: 876,374 | bf16_mfu: 53.83 | total time: 99.27m | eta: 11.5m step 04992/05568 (89.66%) | loss: 2.412262 | lrm: 0.20 | dt: 1189.85ms | tok/sec: 881,268 | bf16_mfu: 54.13 | total time: 99.29m | eta: 11.5m step 04993/05568 (89.67%) | loss: 2.420719 | lrm: 0.20 | dt: 1193.70ms | tok/sec: 878,425 | bf16_mfu: 53.96 | total time: 99.31m | eta: 11.4m step 04994/05568 (89.69%) | loss: 2.419280 | lrm: 0.20 | dt: 1193.17ms | tok/sec: 878,811 | bf16_mfu: 53.98 | total time: 99.33m | eta: 11.4m step 04995/05568 (89.71%) | loss: 2.420511 | lrm: 0.20 | dt: 1197.93ms | tok/sec: 875,320 | bf16_mfu: 53.77 | total time: 99.35m | eta: 11.4m step 04996/05568 (89.73%) | loss: 2.404842 | lrm: 0.20 | dt: 1195.71ms | tok/sec: 876,951 | bf16_mfu: 53.87 | total time: 99.37m | eta: 11.4m step 04997/05568 (89.74%) | loss: 2.413436 | lrm: 0.20 | dt: 1194.75ms | tok/sec: 877,654 | bf16_mfu: 53.91 | total time: 99.39m | eta: 11.4m step 04998/05568 (89.76%) | loss: 2.416921 | lrm: 0.20 | dt: 1197.72ms | tok/sec: 875,474 | bf16_mfu: 53.78 | total time: 99.41m | eta: 11.3m step 04999/05568 (89.78%) | loss: 2.417286 | lrm: 0.20 | dt: 1195.96ms | tok/sec: 876,766 | bf16_mfu: 53.86 | total time: 99.43m | eta: 11.3m step:5000/5568 val_bpb:0.726908 val_time:3.89s step 05000/05568 (89.80%) | loss: 2.404177 | lrm: 0.20 | dt: 1195.04ms | tok/sec: 877,443 | bf16_mfu: 53.90 | total time: 99.45m | eta: 11.3m step 05001/05568 (89.82%) | loss: 2.406452 | lrm: 0.20 | dt: 1163.34ms | tok/sec: 901,348 | bf16_mfu: 55.37 | total time: 99.47m | eta: 11.3m step 05002/05568 (89.83%) | loss: 2.405170 | lrm: 0.20 | dt: 1187.53ms | tok/sec: 882,987 | bf16_mfu: 54.24 | total time: 99.49m | eta: 11.3m step 05003/05568 (89.85%) | loss: 2.401564 | lrm: 0.20 | dt: 1190.56ms | tok/sec: 880,741 | bf16_mfu: 54.10 | total time: 99.51m | eta: 11.2m step 05004/05568 (89.87%) | loss: 2.403738 | lrm: 0.20 | dt: 1192.09ms | tok/sec: 879,610 | bf16_mfu: 54.03 | total time: 99.53m | eta: 11.2m step 05005/05568 (89.89%) | loss: 2.404551 | lrm: 0.20 | dt: 1193.51ms | tok/sec: 878,567 | bf16_mfu: 53.97 | total time: 99.55m | eta: 11.2m step 05006/05568 (89.91%) | loss: 2.402906 | lrm: 0.20 | dt: 1195.58ms | tok/sec: 877,045 | bf16_mfu: 53.87 | total time: 99.57m | eta: 11.2m step 05007/05568 (89.92%) | loss: 2.404788 | lrm: 0.20 | dt: 1190.81ms | tok/sec: 880,557 | bf16_mfu: 54.09 | total time: 99.59m | eta: 11.2m step 05008/05568 (89.94%) | loss: 2.401175 | lrm: 0.20 | dt: 1191.47ms | tok/sec: 880,067 | bf16_mfu: 54.06 | total time: 99.61m | eta: 11.1m step 05009/05568 (89.96%) | loss: 2.410635 | lrm: 0.20 | dt: 1192.02ms | tok/sec: 879,664 | bf16_mfu: 54.04 | total time: 99.63m | eta: 11.1m step 05010/05568 (89.98%) | loss: 2.409028 | lrm: 0.20 | dt: 1191.96ms | tok/sec: 879,704 | bf16_mfu: 54.04 | total time: 99.65m | eta: 11.1m step 05011/05568 (90.00%) | loss: 2.416018 | lrm: 0.20 | dt: 1191.54ms | tok/sec: 880,017 | bf16_mfu: 54.06 | total time: 99.67m | eta: 11.1m step 05012/05568 (90.01%) | loss: 2.408357 | lrm: 0.20 | dt: 1192.14ms | tok/sec: 879,571 | bf16_mfu: 54.03 | total time: 99.69m | eta: 11.1m step 05013/05568 (90.03%) | loss: 2.404609 | lrm: 0.20 | dt: 1191.95ms | tok/sec: 879,717 | bf16_mfu: 54.04 | total time: 99.71m | eta: 11.0m step 05014/05568 (90.05%) | loss: 2.395786 | lrm: 0.20 | dt: 1194.53ms | tok/sec: 877,815 | bf16_mfu: 53.92 | total time: 99.73m | eta: 11.0m step 05015/05568 (90.07%) | loss: 2.391545 | lrm: 0.20 | dt: 1196.02ms | tok/sec: 876,724 | bf16_mfu: 53.86 | total time: 99.75m | eta: 11.0m step 05016/05568 (90.09%) | loss: 2.395049 | lrm: 0.19 | dt: 1192.90ms | tok/sec: 879,015 | bf16_mfu: 54.00 | total time: 99.77m | eta: 11.0m step 05017/05568 (90.10%) | loss: 2.390777 | lrm: 0.19 | dt: 1191.42ms | tok/sec: 880,107 | bf16_mfu: 54.06 | total time: 99.79m | eta: 11.0m step 05018/05568 (90.12%) | loss: 2.391380 | lrm: 0.19 | dt: 1192.53ms | tok/sec: 879,286 | bf16_mfu: 54.01 | total time: 99.81m | eta: 10.9m step 05019/05568 (90.14%) | loss: 2.381295 | lrm: 0.19 | dt: 1192.84ms | tok/sec: 879,056 | bf16_mfu: 54.00 | total time: 99.83m | eta: 10.9m step 05020/05568 (90.16%) | loss: 2.379677 | lrm: 0.19 | dt: 1193.88ms | tok/sec: 878,292 | bf16_mfu: 53.95 | total time: 99.85m | eta: 10.9m step 05021/05568 (90.18%) | loss: 2.383007 | lrm: 0.19 | dt: 1191.09ms | tok/sec: 880,346 | bf16_mfu: 54.08 | total time: 99.87m | eta: 10.9m step 05022/05568 (90.19%) | loss: 2.384092 | lrm: 0.19 | dt: 1196.45ms | tok/sec: 876,409 | bf16_mfu: 53.84 | total time: 99.89m | eta: 10.9m step 05023/05568 (90.21%) | loss: 2.388087 | lrm: 0.19 | dt: 1193.71ms | tok/sec: 878,417 | bf16_mfu: 53.96 | total time: 99.91m | eta: 10.8m step 05024/05568 (90.23%) | loss: 2.388639 | lrm: 0.19 | dt: 1197.04ms | tok/sec: 875,974 | bf16_mfu: 53.81 | total time: 99.93m | eta: 10.8m step 05025/05568 (90.25%) | loss: 2.397914 | lrm: 0.19 | dt: 1193.28ms | tok/sec: 878,732 | bf16_mfu: 53.98 | total time: 99.95m | eta: 10.8m step 05026/05568 (90.27%) | loss: 2.403135 | lrm: 0.19 | dt: 1195.91ms | tok/sec: 876,803 | bf16_mfu: 53.86 | total time: 99.97m | eta: 10.8m step 05027/05568 (90.28%) | loss: 2.407300 | lrm: 0.19 | dt: 1196.82ms | tok/sec: 876,133 | bf16_mfu: 53.82 | total time: 99.99m | eta: 10.8m step 05028/05568 (90.30%) | loss: 2.392979 | lrm: 0.19 | dt: 1195.79ms | tok/sec: 876,887 | bf16_mfu: 53.87 | total time: 100.01m | eta: 10.7m step 05029/05568 (90.32%) | loss: 2.394396 | lrm: 0.19 | dt: 1196.49ms | tok/sec: 876,376 | bf16_mfu: 53.83 | total time: 100.03m | eta: 10.7m step 05030/05568 (90.34%) | loss: 2.384573 | lrm: 0.19 | dt: 1195.14ms | tok/sec: 877,364 | bf16_mfu: 53.89 | total time: 100.05m | eta: 10.7m step 05031/05568 (90.36%) | loss: 2.376670 | lrm: 0.19 | dt: 1192.45ms | tok/sec: 879,345 | bf16_mfu: 54.02 | total time: 100.07m | eta: 10.7m step 05032/05568 (90.37%) | loss: 2.381763 | lrm: 0.19 | dt: 1196.72ms | tok/sec: 876,211 | bf16_mfu: 53.82 | total time: 100.09m | eta: 10.7m step 05033/05568 (90.39%) | loss: 2.364929 | lrm: 0.19 | dt: 1194.28ms | tok/sec: 877,997 | bf16_mfu: 53.93 | total time: 100.11m | eta: 10.6m step 05034/05568 (90.41%) | loss: 2.367462 | lrm: 0.19 | dt: 1193.04ms | tok/sec: 878,910 | bf16_mfu: 53.99 | total time: 100.13m | eta: 10.6m step 05035/05568 (90.43%) | loss: 2.374899 | lrm: 0.19 | dt: 1193.47ms | tok/sec: 878,592 | bf16_mfu: 53.97 | total time: 100.15m | eta: 10.6m step 05036/05568 (90.45%) | loss: 2.382526 | lrm: 0.19 | dt: 1196.31ms | tok/sec: 876,506 | bf16_mfu: 53.84 | total time: 100.17m | eta: 10.6m step 05037/05568 (90.46%) | loss: 2.388051 | lrm: 0.19 | dt: 1195.93ms | tok/sec: 876,785 | bf16_mfu: 53.86 | total time: 100.19m | eta: 10.6m step 05038/05568 (90.48%) | loss: 2.378956 | lrm: 0.19 | dt: 1194.83ms | tok/sec: 877,591 | bf16_mfu: 53.91 | total time: 100.21m | eta: 10.5m step 05039/05568 (90.50%) | loss: 2.378970 | lrm: 0.19 | dt: 1191.52ms | tok/sec: 880,034 | bf16_mfu: 54.06 | total time: 100.23m | eta: 10.5m step 05040/05568 (90.52%) | loss: 2.376358 | lrm: 0.19 | dt: 1194.66ms | tok/sec: 877,720 | bf16_mfu: 53.92 | total time: 100.25m | eta: 10.5m step 05041/05568 (90.54%) | loss: 2.376548 | lrm: 0.19 | dt: 1193.08ms | tok/sec: 878,881 | bf16_mfu: 53.99 | total time: 100.27m | eta: 10.5m step 05042/05568 (90.55%) | loss: 2.382444 | lrm: 0.19 | dt: 1196.36ms | tok/sec: 876,472 | bf16_mfu: 53.84 | total time: 100.29m | eta: 10.5m step 05043/05568 (90.57%) | loss: 2.385059 | lrm: 0.19 | dt: 1193.40ms | tok/sec: 878,647 | bf16_mfu: 53.97 | total time: 100.31m | eta: 10.4m step 05044/05568 (90.59%) | loss: 2.390153 | lrm: 0.19 | dt: 1194.93ms | tok/sec: 877,522 | bf16_mfu: 53.90 | total time: 100.33m | eta: 10.4m step 05045/05568 (90.61%) | loss: 2.376681 | lrm: 0.19 | dt: 1193.75ms | tok/sec: 878,384 | bf16_mfu: 53.96 | total time: 100.35m | eta: 10.4m step 05046/05568 (90.62%) | loss: 2.375682 | lrm: 0.19 | dt: 1196.32ms | tok/sec: 876,504 | bf16_mfu: 53.84 | total time: 100.37m | eta: 10.4m step 05047/05568 (90.64%) | loss: 2.384539 | lrm: 0.19 | dt: 1198.77ms | tok/sec: 874,713 | bf16_mfu: 53.73 | total time: 100.39m | eta: 10.4m step 05048/05568 (90.66%) | loss: 2.399925 | lrm: 0.19 | dt: 1196.07ms | tok/sec: 876,687 | bf16_mfu: 53.85 | total time: 100.41m | eta: 10.3m step 05049/05568 (90.68%) | loss: 2.400993 | lrm: 0.19 | dt: 1192.95ms | tok/sec: 878,975 | bf16_mfu: 53.99 | total time: 100.43m | eta: 10.3m step 05050/05568 (90.70%) | loss: 2.405374 | lrm: 0.19 | dt: 1191.90ms | tok/sec: 879,751 | bf16_mfu: 54.04 | total time: 100.45m | eta: 10.3m step 05051/05568 (90.71%) | loss: 2.393986 | lrm: 0.19 | dt: 1192.88ms | tok/sec: 879,029 | bf16_mfu: 54.00 | total time: 100.47m | eta: 10.3m step 05052/05568 (90.73%) | loss: 2.398145 | lrm: 0.19 | dt: 1188.77ms | tok/sec: 882,066 | bf16_mfu: 54.18 | total time: 100.49m | eta: 10.3m step 05053/05568 (90.75%) | loss: 2.419753 | lrm: 0.19 | dt: 1199.23ms | tok/sec: 874,374 | bf16_mfu: 53.71 | total time: 100.51m | eta: 10.2m step 05054/05568 (90.77%) | loss: 2.420903 | lrm: 0.18 | dt: 1190.77ms | tok/sec: 880,583 | bf16_mfu: 54.09 | total time: 100.53m | eta: 10.2m step 05055/05568 (90.79%) | loss: 2.409759 | lrm: 0.18 | dt: 1190.98ms | tok/sec: 880,433 | bf16_mfu: 54.08 | total time: 100.55m | eta: 10.2m step 05056/05568 (90.80%) | loss: 2.406988 | lrm: 0.18 | dt: 1196.59ms | tok/sec: 876,302 | bf16_mfu: 53.83 | total time: 100.57m | eta: 10.2m step 05057/05568 (90.82%) | loss: 2.401169 | lrm: 0.18 | dt: 1196.13ms | tok/sec: 876,638 | bf16_mfu: 53.85 | total time: 100.59m | eta: 10.2m step 05058/05568 (90.84%) | loss: 2.410816 | lrm: 0.18 | dt: 1195.02ms | tok/sec: 877,451 | bf16_mfu: 53.90 | total time: 100.61m | eta: 10.1m step 05059/05568 (90.86%) | loss: 2.397424 | lrm: 0.18 | dt: 1194.98ms | tok/sec: 877,484 | bf16_mfu: 53.90 | total time: 100.63m | eta: 10.1m step 05060/05568 (90.88%) | loss: 2.398606 | lrm: 0.18 | dt: 1196.25ms | tok/sec: 876,550 | bf16_mfu: 53.84 | total time: 100.65m | eta: 10.1m step 05061/05568 (90.89%) | loss: 2.400209 | lrm: 0.18 | dt: 1190.12ms | tok/sec: 881,070 | bf16_mfu: 54.12 | total time: 100.67m | eta: 10.1m step 05062/05568 (90.91%) | loss: 2.400129 | lrm: 0.18 | dt: 1197.53ms | tok/sec: 875,612 | bf16_mfu: 53.79 | total time: 100.69m | eta: 10.1m step 05063/05568 (90.93%) | loss: 2.401983 | lrm: 0.18 | dt: 1194.65ms | tok/sec: 877,726 | bf16_mfu: 53.92 | total time: 100.71m | eta: 10.0m step 05064/05568 (90.95%) | loss: 2.406427 | lrm: 0.18 | dt: 1191.90ms | tok/sec: 879,752 | bf16_mfu: 54.04 | total time: 100.73m | eta: 10.0m step 05065/05568 (90.97%) | loss: 2.404256 | lrm: 0.18 | dt: 1193.92ms | tok/sec: 878,262 | bf16_mfu: 53.95 | total time: 100.75m | eta: 10.0m step 05066/05568 (90.98%) | loss: 2.415508 | lrm: 0.18 | dt: 1196.95ms | tok/sec: 876,037 | bf16_mfu: 53.81 | total time: 100.77m | eta: 10.0m step 05067/05568 (91.00%) | loss: 2.400705 | lrm: 0.18 | dt: 1191.35ms | tok/sec: 880,156 | bf16_mfu: 54.07 | total time: 100.78m | eta: 10.0m step 05068/05568 (91.02%) | loss: 2.399042 | lrm: 0.18 | dt: 1192.06ms | tok/sec: 879,630 | bf16_mfu: 54.03 | total time: 100.80m | eta: 9.9m step 05069/05568 (91.04%) | loss: 2.396225 | lrm: 0.18 | dt: 1195.95ms | tok/sec: 876,769 | bf16_mfu: 53.86 | total time: 100.82m | eta: 9.9m step 05070/05568 (91.06%) | loss: 2.396447 | lrm: 0.18 | dt: 1200.27ms | tok/sec: 873,614 | bf16_mfu: 53.66 | total time: 100.84m | eta: 9.9m step 05071/05568 (91.07%) | loss: 2.401295 | lrm: 0.18 | dt: 1194.17ms | tok/sec: 878,082 | bf16_mfu: 53.94 | total time: 100.86m | eta: 9.9m step 05072/05568 (91.09%) | loss: 2.409215 | lrm: 0.18 | dt: 1191.39ms | tok/sec: 880,125 | bf16_mfu: 54.06 | total time: 100.88m | eta: 9.9m step 05073/05568 (91.11%) | loss: 2.409234 | lrm: 0.18 | dt: 1196.48ms | tok/sec: 876,386 | bf16_mfu: 53.83 | total time: 100.90m | eta: 9.8m step 05074/05568 (91.13%) | loss: 2.409340 | lrm: 0.18 | dt: 1191.49ms | tok/sec: 880,055 | bf16_mfu: 54.06 | total time: 100.92m | eta: 9.8m step 05075/05568 (91.15%) | loss: 2.405862 | lrm: 0.18 | dt: 1189.92ms | tok/sec: 881,216 | bf16_mfu: 54.13 | total time: 100.94m | eta: 9.8m step 05076/05568 (91.16%) | loss: 2.407823 | lrm: 0.18 | dt: 1196.71ms | tok/sec: 876,212 | bf16_mfu: 53.82 | total time: 100.96m | eta: 9.8m step 05077/05568 (91.18%) | loss: 2.412509 | lrm: 0.18 | dt: 1191.44ms | tok/sec: 880,092 | bf16_mfu: 54.06 | total time: 100.98m | eta: 9.8m step 05078/05568 (91.20%) | loss: 2.416794 | lrm: 0.18 | dt: 1191.63ms | tok/sec: 879,952 | bf16_mfu: 54.05 | total time: 101.00m | eta: 9.7m step 05079/05568 (91.22%) | loss: 2.411633 | lrm: 0.18 | dt: 1191.54ms | tok/sec: 880,014 | bf16_mfu: 54.06 | total time: 101.02m | eta: 9.7m step 05080/05568 (91.24%) | loss: 2.419225 | lrm: 0.18 | dt: 1191.54ms | tok/sec: 880,019 | bf16_mfu: 54.06 | total time: 101.04m | eta: 9.7m step 05081/05568 (91.25%) | loss: 2.425107 | lrm: 0.18 | dt: 1197.02ms | tok/sec: 875,991 | bf16_mfu: 53.81 | total time: 101.06m | eta: 9.7m step 05082/05568 (91.27%) | loss: 2.412451 | lrm: 0.18 | dt: 1194.76ms | tok/sec: 877,647 | bf16_mfu: 53.91 | total time: 101.08m | eta: 9.7m step 05083/05568 (91.29%) | loss: 2.401641 | lrm: 0.18 | dt: 1192.16ms | tok/sec: 879,560 | bf16_mfu: 54.03 | total time: 101.10m | eta: 9.6m step 05084/05568 (91.31%) | loss: 2.398976 | lrm: 0.18 | dt: 1193.37ms | tok/sec: 878,668 | bf16_mfu: 53.97 | total time: 101.12m | eta: 9.6m step 05085/05568 (91.33%) | loss: 2.396052 | lrm: 0.18 | dt: 1192.50ms | tok/sec: 879,306 | bf16_mfu: 54.01 | total time: 101.14m | eta: 9.6m step 05086/05568 (91.34%) | loss: 2.391610 | lrm: 0.18 | dt: 1192.36ms | tok/sec: 879,410 | bf16_mfu: 54.02 | total time: 101.16m | eta: 9.6m step 05087/05568 (91.36%) | loss: 2.378392 | lrm: 0.18 | dt: 1192.24ms | tok/sec: 879,498 | bf16_mfu: 54.03 | total time: 101.18m | eta: 9.6m step 05088/05568 (91.38%) | loss: 2.391541 | lrm: 0.18 | dt: 1188.81ms | tok/sec: 882,040 | bf16_mfu: 54.18 | total time: 101.20m | eta: 9.5m step 05089/05568 (91.40%) | loss: 2.399832 | lrm: 0.18 | dt: 1192.67ms | tok/sec: 879,183 | bf16_mfu: 54.01 | total time: 101.22m | eta: 9.5m step 05090/05568 (91.42%) | loss: 2.402618 | lrm: 0.18 | dt: 1197.67ms | tok/sec: 875,510 | bf16_mfu: 53.78 | total time: 101.24m | eta: 9.5m step 05091/05568 (91.43%) | loss: 2.401665 | lrm: 0.18 | dt: 1193.63ms | tok/sec: 878,478 | bf16_mfu: 53.96 | total time: 101.26m | eta: 9.5m step 05092/05568 (91.45%) | loss: 2.399366 | lrm: 0.17 | dt: 1193.54ms | tok/sec: 878,544 | bf16_mfu: 53.97 | total time: 101.28m | eta: 9.5m step 05093/05568 (91.47%) | loss: 2.401464 | lrm: 0.17 | dt: 1192.83ms | tok/sec: 879,067 | bf16_mfu: 54.00 | total time: 101.30m | eta: 9.4m step 05094/05568 (91.49%) | loss: 2.375842 | lrm: 0.17 | dt: 1196.00ms | tok/sec: 876,734 | bf16_mfu: 53.86 | total time: 101.32m | eta: 9.4m step 05095/05568 (91.51%) | loss: 2.373418 | lrm: 0.17 | dt: 1189.31ms | tok/sec: 881,667 | bf16_mfu: 54.16 | total time: 101.34m | eta: 9.4m step 05096/05568 (91.52%) | loss: 2.384625 | lrm: 0.17 | dt: 1193.93ms | tok/sec: 878,252 | bf16_mfu: 53.95 | total time: 101.36m | eta: 9.4m step 05097/05568 (91.54%) | loss: 2.380331 | lrm: 0.17 | dt: 1191.04ms | tok/sec: 880,387 | bf16_mfu: 54.08 | total time: 101.38m | eta: 9.4m step 05098/05568 (91.56%) | loss: 2.369335 | lrm: 0.17 | dt: 1194.42ms | tok/sec: 877,895 | bf16_mfu: 53.93 | total time: 101.40m | eta: 9.3m step 05099/05568 (91.58%) | loss: 2.370702 | lrm: 0.17 | dt: 1192.85ms | tok/sec: 879,050 | bf16_mfu: 54.00 | total time: 101.42m | eta: 9.3m step 05100/05568 (91.59%) | loss: 2.373832 | lrm: 0.17 | dt: 1193.56ms | tok/sec: 878,525 | bf16_mfu: 53.97 | total time: 101.44m | eta: 9.3m step 05101/05568 (91.61%) | loss: 2.363081 | lrm: 0.17 | dt: 1195.24ms | tok/sec: 877,295 | bf16_mfu: 53.89 | total time: 101.46m | eta: 9.3m step 05102/05568 (91.63%) | loss: 2.371685 | lrm: 0.17 | dt: 1190.76ms | tok/sec: 880,596 | bf16_mfu: 54.09 | total time: 101.48m | eta: 9.3m step 05103/05568 (91.65%) | loss: 2.385592 | lrm: 0.17 | dt: 1190.87ms | tok/sec: 880,515 | bf16_mfu: 54.09 | total time: 101.50m | eta: 9.2m step 05104/05568 (91.67%) | loss: 2.380084 | lrm: 0.17 | dt: 1189.47ms | tok/sec: 881,547 | bf16_mfu: 54.15 | total time: 101.52m | eta: 9.2m step 05105/05568 (91.68%) | loss: 2.388787 | lrm: 0.17 | dt: 1194.60ms | tok/sec: 877,762 | bf16_mfu: 53.92 | total time: 101.54m | eta: 9.2m step 05106/05568 (91.70%) | loss: 2.387676 | lrm: 0.17 | dt: 1191.13ms | tok/sec: 880,322 | bf16_mfu: 54.08 | total time: 101.56m | eta: 9.2m step 05107/05568 (91.72%) | loss: 2.385049 | lrm: 0.17 | dt: 1196.36ms | tok/sec: 876,471 | bf16_mfu: 53.84 | total time: 101.58m | eta: 9.2m step 05108/05568 (91.74%) | loss: 2.388428 | lrm: 0.17 | dt: 1192.82ms | tok/sec: 879,075 | bf16_mfu: 54.00 | total time: 101.60m | eta: 9.1m step 05109/05568 (91.76%) | loss: 2.393048 | lrm: 0.17 | dt: 1190.74ms | tok/sec: 880,612 | bf16_mfu: 54.09 | total time: 101.62m | eta: 9.1m step 05110/05568 (91.77%) | loss: 2.391106 | lrm: 0.17 | dt: 1195.01ms | tok/sec: 877,460 | bf16_mfu: 53.90 | total time: 101.64m | eta: 9.1m step 05111/05568 (91.79%) | loss: 2.389378 | lrm: 0.17 | dt: 1201.50ms | tok/sec: 872,720 | bf16_mfu: 53.61 | total time: 101.66m | eta: 9.1m step 05112/05568 (91.81%) | loss: 2.393695 | lrm: 0.17 | dt: 1191.19ms | tok/sec: 880,273 | bf16_mfu: 54.07 | total time: 101.68m | eta: 9.1m step 05113/05568 (91.83%) | loss: 2.400540 | lrm: 0.17 | dt: 1194.59ms | tok/sec: 877,770 | bf16_mfu: 53.92 | total time: 101.70m | eta: 9.0m step 05114/05568 (91.85%) | loss: 2.400719 | lrm: 0.17 | dt: 1190.41ms | tok/sec: 880,854 | bf16_mfu: 54.11 | total time: 101.72m | eta: 9.0m step 05115/05568 (91.86%) | loss: 2.391669 | lrm: 0.17 | dt: 1190.94ms | tok/sec: 880,463 | bf16_mfu: 54.08 | total time: 101.74m | eta: 9.0m step 05116/05568 (91.88%) | loss: 2.391598 | lrm: 0.17 | dt: 1192.46ms | tok/sec: 879,337 | bf16_mfu: 54.02 | total time: 101.76m | eta: 9.0m step 05117/05568 (91.90%) | loss: 2.394697 | lrm: 0.17 | dt: 1200.11ms | tok/sec: 873,735 | bf16_mfu: 53.67 | total time: 101.78m | eta: 9.0m step 05118/05568 (91.92%) | loss: 2.403432 | lrm: 0.17 | dt: 1194.38ms | tok/sec: 877,922 | bf16_mfu: 53.93 | total time: 101.80m | eta: 8.9m step 05119/05568 (91.94%) | loss: 2.400035 | lrm: 0.17 | dt: 1193.55ms | tok/sec: 878,534 | bf16_mfu: 53.97 | total time: 101.82m | eta: 8.9m step 05120/05568 (91.95%) | loss: 2.412804 | lrm: 0.17 | dt: 1192.10ms | tok/sec: 879,604 | bf16_mfu: 54.03 | total time: 101.84m | eta: 8.9m step 05121/05568 (91.97%) | loss: 2.413445 | lrm: 0.17 | dt: 1194.91ms | tok/sec: 877,538 | bf16_mfu: 53.91 | total time: 101.86m | eta: 8.9m step 05122/05568 (91.99%) | loss: 2.406872 | lrm: 0.17 | dt: 1191.67ms | tok/sec: 879,920 | bf16_mfu: 54.05 | total time: 101.88m | eta: 8.9m step 05123/05568 (92.01%) | loss: 2.401790 | lrm: 0.17 | dt: 1196.24ms | tok/sec: 876,559 | bf16_mfu: 53.85 | total time: 101.90m | eta: 8.8m step 05124/05568 (92.03%) | loss: 2.394194 | lrm: 0.17 | dt: 1191.70ms | tok/sec: 879,898 | bf16_mfu: 54.05 | total time: 101.92m | eta: 8.8m step 05125/05568 (92.04%) | loss: 2.408493 | lrm: 0.17 | dt: 1194.63ms | tok/sec: 877,740 | bf16_mfu: 53.92 | total time: 101.94m | eta: 8.8m step 05126/05568 (92.06%) | loss: 2.392705 | lrm: 0.17 | dt: 1195.15ms | tok/sec: 877,361 | bf16_mfu: 53.89 | total time: 101.96m | eta: 8.8m step 05127/05568 (92.08%) | loss: 2.391396 | lrm: 0.17 | dt: 1188.52ms | tok/sec: 882,250 | bf16_mfu: 54.19 | total time: 101.98m | eta: 8.8m step 05128/05568 (92.10%) | loss: 2.380963 | lrm: 0.17 | dt: 1192.33ms | tok/sec: 879,434 | bf16_mfu: 54.02 | total time: 102.00m | eta: 8.7m step 05129/05568 (92.12%) | loss: 2.373254 | lrm: 0.17 | dt: 1191.49ms | tok/sec: 880,051 | bf16_mfu: 54.06 | total time: 102.02m | eta: 8.7m step 05130/05568 (92.13%) | loss: 2.369210 | lrm: 0.16 | dt: 1194.99ms | tok/sec: 877,473 | bf16_mfu: 53.90 | total time: 102.04m | eta: 8.7m step 05131/05568 (92.15%) | loss: 2.377689 | lrm: 0.16 | dt: 1192.07ms | tok/sec: 879,624 | bf16_mfu: 54.03 | total time: 102.06m | eta: 8.7m step 05132/05568 (92.17%) | loss: 2.375587 | lrm: 0.16 | dt: 1191.30ms | tok/sec: 880,191 | bf16_mfu: 54.07 | total time: 102.08m | eta: 8.7m step 05133/05568 (92.19%) | loss: 2.368624 | lrm: 0.16 | dt: 1195.30ms | tok/sec: 877,252 | bf16_mfu: 53.89 | total time: 102.10m | eta: 8.6m step 05134/05568 (92.21%) | loss: 2.371986 | lrm: 0.16 | dt: 1192.04ms | tok/sec: 879,651 | bf16_mfu: 54.03 | total time: 102.12m | eta: 8.6m step 05135/05568 (92.22%) | loss: 2.372380 | lrm: 0.16 | dt: 1191.91ms | tok/sec: 879,742 | bf16_mfu: 54.04 | total time: 102.14m | eta: 8.6m step 05136/05568 (92.24%) | loss: 2.368509 | lrm: 0.16 | dt: 1193.35ms | tok/sec: 878,682 | bf16_mfu: 53.98 | total time: 102.16m | eta: 8.6m step 05137/05568 (92.26%) | loss: 2.363430 | lrm: 0.16 | dt: 1195.97ms | tok/sec: 876,757 | bf16_mfu: 53.86 | total time: 102.18m | eta: 8.6m step 05138/05568 (92.28%) | loss: 2.371842 | lrm: 0.16 | dt: 1192.75ms | tok/sec: 879,124 | bf16_mfu: 54.00 | total time: 102.20m | eta: 8.5m step 05139/05568 (92.30%) | loss: 2.367012 | lrm: 0.16 | dt: 1196.73ms | tok/sec: 876,199 | bf16_mfu: 53.82 | total time: 102.22m | eta: 8.5m step 05140/05568 (92.31%) | loss: 2.366016 | lrm: 0.16 | dt: 1194.47ms | tok/sec: 877,860 | bf16_mfu: 53.92 | total time: 102.24m | eta: 8.5m step 05141/05568 (92.33%) | loss: 2.357658 | lrm: 0.16 | dt: 1197.49ms | tok/sec: 875,643 | bf16_mfu: 53.79 | total time: 102.26m | eta: 8.5m step 05142/05568 (92.35%) | loss: 2.367109 | lrm: 0.16 | dt: 1190.87ms | tok/sec: 880,510 | bf16_mfu: 54.09 | total time: 102.28m | eta: 8.5m step 05143/05568 (92.37%) | loss: 2.357137 | lrm: 0.16 | dt: 1195.94ms | tok/sec: 876,779 | bf16_mfu: 53.86 | total time: 102.30m | eta: 8.4m step 05144/05568 (92.39%) | loss: 2.363490 | lrm: 0.16 | dt: 1192.94ms | tok/sec: 878,983 | bf16_mfu: 53.99 | total time: 102.32m | eta: 8.4m step 05145/05568 (92.40%) | loss: 2.364254 | lrm: 0.16 | dt: 1190.17ms | tok/sec: 881,027 | bf16_mfu: 54.12 | total time: 102.34m | eta: 8.4m step 05146/05568 (92.42%) | loss: 2.368381 | lrm: 0.16 | dt: 1196.48ms | tok/sec: 876,383 | bf16_mfu: 53.83 | total time: 102.36m | eta: 8.4m step 05147/05568 (92.44%) | loss: 2.376067 | lrm: 0.16 | dt: 1196.07ms | tok/sec: 876,681 | bf16_mfu: 53.85 | total time: 102.38m | eta: 8.4m step 05148/05568 (92.46%) | loss: 2.373118 | lrm: 0.16 | dt: 1190.13ms | tok/sec: 881,056 | bf16_mfu: 54.12 | total time: 102.40m | eta: 8.4m step 05149/05568 (92.47%) | loss: 2.380737 | lrm: 0.16 | dt: 1198.56ms | tok/sec: 874,866 | bf16_mfu: 53.74 | total time: 102.42m | eta: 8.3m step 05150/05568 (92.49%) | loss: 2.375857 | lrm: 0.16 | dt: 1197.32ms | tok/sec: 875,765 | bf16_mfu: 53.80 | total time: 102.44m | eta: 8.3m step 05151/05568 (92.51%) | loss: 2.368712 | lrm: 0.16 | dt: 1192.70ms | tok/sec: 879,164 | bf16_mfu: 54.01 | total time: 102.46m | eta: 8.3m step 05152/05568 (92.53%) | loss: 2.358666 | lrm: 0.16 | dt: 1195.83ms | tok/sec: 876,860 | bf16_mfu: 53.86 | total time: 102.48m | eta: 8.3m step 05153/05568 (92.55%) | loss: 2.354789 | lrm: 0.16 | dt: 1190.03ms | tok/sec: 881,136 | bf16_mfu: 54.13 | total time: 102.50m | eta: 8.3m step 05154/05568 (92.56%) | loss: 2.346134 | lrm: 0.16 | dt: 1189.83ms | tok/sec: 881,283 | bf16_mfu: 54.14 | total time: 102.52m | eta: 8.2m step 05155/05568 (92.58%) | loss: 2.341515 | lrm: 0.16 | dt: 1195.84ms | tok/sec: 876,855 | bf16_mfu: 53.86 | total time: 102.54m | eta: 8.2m step 05156/05568 (92.60%) | loss: 2.341836 | lrm: 0.16 | dt: 1191.13ms | tok/sec: 880,317 | bf16_mfu: 54.08 | total time: 102.56m | eta: 8.2m step 05157/05568 (92.62%) | loss: 2.335498 | lrm: 0.16 | dt: 1195.48ms | tok/sec: 877,115 | bf16_mfu: 53.88 | total time: 102.58m | eta: 8.2m step 05158/05568 (92.64%) | loss: 2.320045 | lrm: 0.16 | dt: 1194.34ms | tok/sec: 877,957 | bf16_mfu: 53.93 | total time: 102.60m | eta: 8.2m step 05159/05568 (92.65%) | loss: 2.319964 | lrm: 0.16 | dt: 1188.00ms | tok/sec: 882,640 | bf16_mfu: 54.22 | total time: 102.61m | eta: 8.1m step 05160/05568 (92.67%) | loss: 2.322585 | lrm: 0.16 | dt: 1193.27ms | tok/sec: 878,738 | bf16_mfu: 53.98 | total time: 102.63m | eta: 8.1m step 05161/05568 (92.69%) | loss: 2.322477 | lrm: 0.16 | dt: 1191.94ms | tok/sec: 879,724 | bf16_mfu: 54.04 | total time: 102.65m | eta: 8.1m step 05162/05568 (92.71%) | loss: 2.321354 | lrm: 0.16 | dt: 1194.58ms | tok/sec: 877,774 | bf16_mfu: 53.92 | total time: 102.67m | eta: 8.1m step 05163/05568 (92.73%) | loss: 2.327842 | lrm: 0.16 | dt: 1195.47ms | tok/sec: 877,124 | bf16_mfu: 53.88 | total time: 102.69m | eta: 8.1m step 05164/05568 (92.74%) | loss: 2.326541 | lrm: 0.16 | dt: 1187.79ms | tok/sec: 882,792 | bf16_mfu: 54.23 | total time: 102.71m | eta: 8.0m step 05165/05568 (92.76%) | loss: 2.324170 | lrm: 0.16 | dt: 1193.64ms | tok/sec: 878,469 | bf16_mfu: 53.96 | total time: 102.73m | eta: 8.0m step 05166/05568 (92.78%) | loss: 2.336897 | lrm: 0.16 | dt: 1186.20ms | tok/sec: 883,976 | bf16_mfu: 54.30 | total time: 102.75m | eta: 8.0m step 05167/05568 (92.80%) | loss: 2.352572 | lrm: 0.16 | dt: 1193.55ms | tok/sec: 878,532 | bf16_mfu: 53.97 | total time: 102.77m | eta: 8.0m step 05168/05568 (92.82%) | loss: 2.354572 | lrm: 0.16 | dt: 1193.56ms | tok/sec: 878,525 | bf16_mfu: 53.97 | total time: 102.79m | eta: 8.0m step 05169/05568 (92.83%) | loss: 2.353527 | lrm: 0.15 | dt: 1192.95ms | tok/sec: 878,973 | bf16_mfu: 53.99 | total time: 102.81m | eta: 7.9m step 05170/05568 (92.85%) | loss: 2.361625 | lrm: 0.15 | dt: 1196.54ms | tok/sec: 876,343 | bf16_mfu: 53.83 | total time: 102.83m | eta: 7.9m step 05171/05568 (92.87%) | loss: 2.367342 | lrm: 0.15 | dt: 1192.46ms | tok/sec: 879,337 | bf16_mfu: 54.02 | total time: 102.85m | eta: 7.9m step 05172/05568 (92.89%) | loss: 2.375513 | lrm: 0.15 | dt: 1192.68ms | tok/sec: 879,179 | bf16_mfu: 54.01 | total time: 102.87m | eta: 7.9m step 05173/05568 (92.91%) | loss: 2.359827 | lrm: 0.15 | dt: 1196.09ms | tok/sec: 876,667 | bf16_mfu: 53.85 | total time: 102.89m | eta: 7.9m step 05174/05568 (92.92%) | loss: 2.364000 | lrm: 0.15 | dt: 1194.55ms | tok/sec: 877,801 | bf16_mfu: 53.92 | total time: 102.91m | eta: 7.8m step 05175/05568 (92.94%) | loss: 2.360557 | lrm: 0.15 | dt: 1193.04ms | tok/sec: 878,908 | bf16_mfu: 53.99 | total time: 102.93m | eta: 7.8m step 05176/05568 (92.96%) | loss: 2.371438 | lrm: 0.15 | dt: 1192.26ms | tok/sec: 879,487 | bf16_mfu: 54.02 | total time: 102.95m | eta: 7.8m step 05177/05568 (92.98%) | loss: 2.370870 | lrm: 0.15 | dt: 1192.02ms | tok/sec: 879,661 | bf16_mfu: 54.04 | total time: 102.97m | eta: 7.8m step 05178/05568 (93.00%) | loss: 2.367414 | lrm: 0.15 | dt: 1191.83ms | tok/sec: 879,801 | bf16_mfu: 54.04 | total time: 102.99m | eta: 7.8m step 05179/05568 (93.01%) | loss: 2.372533 | lrm: 0.15 | dt: 1194.13ms | tok/sec: 878,105 | bf16_mfu: 53.94 | total time: 103.01m | eta: 7.7m step 05180/05568 (93.03%) | loss: 2.367780 | lrm: 0.15 | dt: 1191.39ms | tok/sec: 880,130 | bf16_mfu: 54.06 | total time: 103.03m | eta: 7.7m step 05181/05568 (93.05%) | loss: 2.378276 | lrm: 0.15 | dt: 1189.93ms | tok/sec: 881,209 | bf16_mfu: 54.13 | total time: 103.05m | eta: 7.7m step 05182/05568 (93.07%) | loss: 2.372680 | lrm: 0.15 | dt: 1192.58ms | tok/sec: 879,252 | bf16_mfu: 54.01 | total time: 103.07m | eta: 7.7m step 05183/05568 (93.09%) | loss: 2.383453 | lrm: 0.15 | dt: 1191.53ms | tok/sec: 880,025 | bf16_mfu: 54.06 | total time: 103.09m | eta: 7.7m step 05184/05568 (93.10%) | loss: 2.376387 | lrm: 0.15 | dt: 1193.59ms | tok/sec: 878,508 | bf16_mfu: 53.96 | total time: 103.11m | eta: 7.6m step 05185/05568 (93.12%) | loss: 2.382559 | lrm: 0.15 | dt: 1195.08ms | tok/sec: 877,413 | bf16_mfu: 53.90 | total time: 103.13m | eta: 7.6m step 05186/05568 (93.14%) | loss: 2.379922 | lrm: 0.15 | dt: 1195.60ms | tok/sec: 877,027 | bf16_mfu: 53.87 | total time: 103.15m | eta: 7.6m step 05187/05568 (93.16%) | loss: 2.382359 | lrm: 0.15 | dt: 1191.99ms | tok/sec: 879,682 | bf16_mfu: 54.04 | total time: 103.17m | eta: 7.6m step 05188/05568 (93.18%) | loss: 2.385679 | lrm: 0.15 | dt: 1194.07ms | tok/sec: 878,149 | bf16_mfu: 53.94 | total time: 103.19m | eta: 7.6m step 05189/05568 (93.19%) | loss: 2.380855 | lrm: 0.15 | dt: 1197.13ms | tok/sec: 875,905 | bf16_mfu: 53.80 | total time: 103.21m | eta: 7.5m step 05190/05568 (93.21%) | loss: 2.378202 | lrm: 0.15 | dt: 1188.75ms | tok/sec: 882,080 | bf16_mfu: 54.18 | total time: 103.23m | eta: 7.5m step 05191/05568 (93.23%) | loss: 2.379070 | lrm: 0.15 | dt: 1197.26ms | tok/sec: 875,815 | bf16_mfu: 53.80 | total time: 103.25m | eta: 7.5m step 05192/05568 (93.25%) | loss: 2.372310 | lrm: 0.15 | dt: 1191.96ms | tok/sec: 879,705 | bf16_mfu: 54.04 | total time: 103.27m | eta: 7.5m step 05193/05568 (93.27%) | loss: 2.379336 | lrm: 0.15 | dt: 1193.82ms | tok/sec: 878,340 | bf16_mfu: 53.95 | total time: 103.29m | eta: 7.5m step 05194/05568 (93.28%) | loss: 2.391952 | lrm: 0.15 | dt: 1193.50ms | tok/sec: 878,571 | bf16_mfu: 53.97 | total time: 103.31m | eta: 7.4m step 05195/05568 (93.30%) | loss: 2.386609 | lrm: 0.15 | dt: 1196.27ms | tok/sec: 876,537 | bf16_mfu: 53.84 | total time: 103.33m | eta: 7.4m step 05196/05568 (93.32%) | loss: 2.386818 | lrm: 0.15 | dt: 1195.33ms | tok/sec: 877,224 | bf16_mfu: 53.89 | total time: 103.35m | eta: 7.4m step 05197/05568 (93.34%) | loss: 2.389316 | lrm: 0.15 | dt: 1194.72ms | tok/sec: 877,674 | bf16_mfu: 53.91 | total time: 103.37m | eta: 7.4m step 05198/05568 (93.35%) | loss: 2.401931 | lrm: 0.15 | dt: 1198.95ms | tok/sec: 874,579 | bf16_mfu: 53.72 | total time: 103.39m | eta: 7.4m step 05199/05568 (93.37%) | loss: 2.395713 | lrm: 0.15 | dt: 1194.06ms | tok/sec: 878,161 | bf16_mfu: 53.94 | total time: 103.41m | eta: 7.3m step 05200/05568 (93.39%) | loss: 2.398227 | lrm: 0.15 | dt: 1190.60ms | tok/sec: 880,713 | bf16_mfu: 54.10 | total time: 103.43m | eta: 7.3m step 05201/05568 (93.41%) | loss: 2.397603 | lrm: 0.15 | dt: 1195.20ms | tok/sec: 877,321 | bf16_mfu: 53.89 | total time: 103.45m | eta: 7.3m step 05202/05568 (93.43%) | loss: 2.382740 | lrm: 0.15 | dt: 1193.07ms | tok/sec: 878,888 | bf16_mfu: 53.99 | total time: 103.47m | eta: 7.3m step 05203/05568 (93.44%) | loss: 2.374092 | lrm: 0.15 | dt: 1198.51ms | tok/sec: 874,901 | bf16_mfu: 53.74 | total time: 103.49m | eta: 7.3m step 05204/05568 (93.46%) | loss: 2.372291 | lrm: 0.15 | dt: 1189.04ms | tok/sec: 881,866 | bf16_mfu: 54.17 | total time: 103.51m | eta: 7.2m step 05205/05568 (93.48%) | loss: 2.364893 | lrm: 0.15 | dt: 1195.79ms | tok/sec: 876,891 | bf16_mfu: 53.87 | total time: 103.53m | eta: 7.2m step 05206/05568 (93.50%) | loss: 2.380402 | lrm: 0.15 | dt: 1195.21ms | tok/sec: 877,313 | bf16_mfu: 53.89 | total time: 103.55m | eta: 7.2m step 05207/05568 (93.52%) | loss: 2.386314 | lrm: 0.14 | dt: 1196.08ms | tok/sec: 876,673 | bf16_mfu: 53.85 | total time: 103.57m | eta: 7.2m step 05208/05568 (93.53%) | loss: 2.379539 | lrm: 0.14 | dt: 1193.70ms | tok/sec: 878,421 | bf16_mfu: 53.96 | total time: 103.59m | eta: 7.2m step 05209/05568 (93.55%) | loss: 2.374301 | lrm: 0.14 | dt: 1198.20ms | tok/sec: 875,127 | bf16_mfu: 53.76 | total time: 103.61m | eta: 7.1m step 05210/05568 (93.57%) | loss: 2.370304 | lrm: 0.14 | dt: 1193.71ms | tok/sec: 878,415 | bf16_mfu: 53.96 | total time: 103.63m | eta: 7.1m step 05211/05568 (93.59%) | loss: 2.384206 | lrm: 0.14 | dt: 1196.87ms | tok/sec: 876,099 | bf16_mfu: 53.82 | total time: 103.65m | eta: 7.1m step 05212/05568 (93.61%) | loss: 2.382287 | lrm: 0.14 | dt: 1191.22ms | tok/sec: 880,251 | bf16_mfu: 54.07 | total time: 103.67m | eta: 7.1m step 05213/05568 (93.62%) | loss: 2.376762 | lrm: 0.14 | dt: 1194.82ms | tok/sec: 877,599 | bf16_mfu: 53.91 | total time: 103.69m | eta: 7.1m step 05214/05568 (93.64%) | loss: 2.384893 | lrm: 0.14 | dt: 1195.04ms | tok/sec: 877,439 | bf16_mfu: 53.90 | total time: 103.71m | eta: 7.0m step 05215/05568 (93.66%) | loss: 2.389187 | lrm: 0.14 | dt: 1193.18ms | tok/sec: 878,810 | bf16_mfu: 53.98 | total time: 103.73m | eta: 7.0m step 05216/05568 (93.68%) | loss: 2.387021 | lrm: 0.14 | dt: 1192.86ms | tok/sec: 879,040 | bf16_mfu: 54.00 | total time: 103.75m | eta: 7.0m step 05217/05568 (93.70%) | loss: 2.387421 | lrm: 0.14 | dt: 1194.45ms | tok/sec: 877,875 | bf16_mfu: 53.93 | total time: 103.77m | eta: 7.0m step 05218/05568 (93.71%) | loss: 2.388977 | lrm: 0.14 | dt: 1190.67ms | tok/sec: 880,658 | bf16_mfu: 54.10 | total time: 103.79m | eta: 7.0m step 05219/05568 (93.73%) | loss: 2.394793 | lrm: 0.14 | dt: 1194.95ms | tok/sec: 877,505 | bf16_mfu: 53.90 | total time: 103.81m | eta: 6.9m step 05220/05568 (93.75%) | loss: 2.392914 | lrm: 0.14 | dt: 1193.36ms | tok/sec: 878,676 | bf16_mfu: 53.98 | total time: 103.83m | eta: 6.9m step 05221/05568 (93.77%) | loss: 2.373905 | lrm: 0.14 | dt: 1197.19ms | tok/sec: 875,867 | bf16_mfu: 53.80 | total time: 103.85m | eta: 6.9m step 05222/05568 (93.79%) | loss: 2.371060 | lrm: 0.14 | dt: 1195.77ms | tok/sec: 876,902 | bf16_mfu: 53.87 | total time: 103.87m | eta: 6.9m step 05223/05568 (93.80%) | loss: 2.364188 | lrm: 0.14 | dt: 1191.76ms | tok/sec: 879,854 | bf16_mfu: 54.05 | total time: 103.89m | eta: 6.9m step 05224/05568 (93.82%) | loss: 2.356142 | lrm: 0.14 | dt: 1199.19ms | tok/sec: 874,404 | bf16_mfu: 53.71 | total time: 103.91m | eta: 6.8m step 05225/05568 (93.84%) | loss: 2.346741 | lrm: 0.14 | dt: 1193.22ms | tok/sec: 878,775 | bf16_mfu: 53.98 | total time: 103.93m | eta: 6.8m step 05226/05568 (93.86%) | loss: 2.342238 | lrm: 0.14 | dt: 1199.76ms | tok/sec: 873,987 | bf16_mfu: 53.69 | total time: 103.95m | eta: 6.8m step 05227/05568 (93.88%) | loss: 2.352216 | lrm: 0.14 | dt: 1194.72ms | tok/sec: 877,673 | bf16_mfu: 53.91 | total time: 103.97m | eta: 6.8m step 05228/05568 (93.89%) | loss: 2.355153 | lrm: 0.14 | dt: 1198.94ms | tok/sec: 874,588 | bf16_mfu: 53.72 | total time: 103.99m | eta: 6.8m step 05229/05568 (93.91%) | loss: 2.360141 | lrm: 0.14 | dt: 1194.85ms | tok/sec: 877,579 | bf16_mfu: 53.91 | total time: 104.01m | eta: 6.7m step 05230/05568 (93.93%) | loss: 2.359212 | lrm: 0.14 | dt: 1191.18ms | tok/sec: 880,284 | bf16_mfu: 54.07 | total time: 104.03m | eta: 6.7m step 05231/05568 (93.95%) | loss: 2.364965 | lrm: 0.14 | dt: 1191.76ms | tok/sec: 879,852 | bf16_mfu: 54.05 | total time: 104.05m | eta: 6.7m step 05232/05568 (93.97%) | loss: 2.370025 | lrm: 0.14 | dt: 1195.40ms | tok/sec: 877,173 | bf16_mfu: 53.88 | total time: 104.07m | eta: 6.7m step 05233/05568 (93.98%) | loss: 2.369459 | lrm: 0.14 | dt: 1192.93ms | tok/sec: 878,994 | bf16_mfu: 53.99 | total time: 104.09m | eta: 6.7m step 05234/05568 (94.00%) | loss: 2.372828 | lrm: 0.14 | dt: 1195.67ms | tok/sec: 876,975 | bf16_mfu: 53.87 | total time: 104.11m | eta: 6.6m step 05235/05568 (94.02%) | loss: 2.371117 | lrm: 0.14 | dt: 1198.83ms | tok/sec: 874,667 | bf16_mfu: 53.73 | total time: 104.13m | eta: 6.6m step 05236/05568 (94.04%) | loss: 2.368886 | lrm: 0.14 | dt: 1190.05ms | tok/sec: 881,116 | bf16_mfu: 54.12 | total time: 104.15m | eta: 6.6m step 05237/05568 (94.06%) | loss: 2.367171 | lrm: 0.14 | dt: 1195.11ms | tok/sec: 877,389 | bf16_mfu: 53.90 | total time: 104.17m | eta: 6.6m step 05238/05568 (94.07%) | loss: 2.386127 | lrm: 0.14 | dt: 1194.64ms | tok/sec: 877,736 | bf16_mfu: 53.92 | total time: 104.19m | eta: 6.6m step 05239/05568 (94.09%) | loss: 2.375979 | lrm: 0.14 | dt: 1193.72ms | tok/sec: 878,407 | bf16_mfu: 53.96 | total time: 104.21m | eta: 6.5m step 05240/05568 (94.11%) | loss: 2.379269 | lrm: 0.14 | dt: 1194.07ms | tok/sec: 878,152 | bf16_mfu: 53.94 | total time: 104.23m | eta: 6.5m step 05241/05568 (94.13%) | loss: 2.389988 | lrm: 0.14 | dt: 1192.38ms | tok/sec: 879,396 | bf16_mfu: 54.02 | total time: 104.25m | eta: 6.5m step 05242/05568 (94.15%) | loss: 2.396445 | lrm: 0.14 | dt: 1192.13ms | tok/sec: 879,579 | bf16_mfu: 54.03 | total time: 104.27m | eta: 6.5m step 05243/05568 (94.16%) | loss: 2.393825 | lrm: 0.14 | dt: 1196.41ms | tok/sec: 876,432 | bf16_mfu: 53.84 | total time: 104.29m | eta: 6.5m step 05244/05568 (94.18%) | loss: 2.383828 | lrm: 0.14 | dt: 1197.30ms | tok/sec: 875,781 | bf16_mfu: 53.80 | total time: 104.31m | eta: 6.4m step 05245/05568 (94.20%) | loss: 2.375191 | lrm: 0.13 | dt: 1194.04ms | tok/sec: 878,178 | bf16_mfu: 53.94 | total time: 104.33m | eta: 6.4m step 05246/05568 (94.22%) | loss: 2.384764 | lrm: 0.13 | dt: 1194.40ms | tok/sec: 877,908 | bf16_mfu: 53.93 | total time: 104.35m | eta: 6.4m step 05247/05568 (94.23%) | loss: 2.383460 | lrm: 0.13 | dt: 1194.54ms | tok/sec: 877,807 | bf16_mfu: 53.92 | total time: 104.37m | eta: 6.4m step 05248/05568 (94.25%) | loss: 2.393825 | lrm: 0.13 | dt: 1195.95ms | tok/sec: 876,773 | bf16_mfu: 53.86 | total time: 104.39m | eta: 6.4m step 05249/05568 (94.27%) | loss: 2.396282 | lrm: 0.13 | dt: 1196.03ms | tok/sec: 876,716 | bf16_mfu: 53.85 | total time: 104.41m | eta: 6.3m step:5250/5568 val_bpb:0.722909 val_time:3.90s step 05250/05568 (94.29%) | loss: 2.391108 | lrm: 0.13 | dt: 1197.86ms | tok/sec: 875,373 | bf16_mfu: 53.77 | total time: 104.43m | eta: 6.3m step 05251/05568 (94.31%) | loss: 2.388433 | lrm: 0.13 | dt: 1191.86ms | tok/sec: 879,783 | bf16_mfu: 54.04 | total time: 104.45m | eta: 6.3m step 05252/05568 (94.32%) | loss: 2.381432 | lrm: 0.13 | dt: 1192.69ms | tok/sec: 879,169 | bf16_mfu: 54.01 | total time: 104.47m | eta: 6.3m step 05253/05568 (94.34%) | loss: 2.392623 | lrm: 0.13 | dt: 1189.47ms | tok/sec: 881,550 | bf16_mfu: 54.15 | total time: 104.49m | eta: 6.3m step 05254/05568 (94.36%) | loss: 2.384944 | lrm: 0.13 | dt: 1192.08ms | tok/sec: 879,619 | bf16_mfu: 54.03 | total time: 104.51m | eta: 6.2m step 05255/05568 (94.38%) | loss: 2.388371 | lrm: 0.13 | dt: 1196.66ms | tok/sec: 876,255 | bf16_mfu: 53.83 | total time: 104.53m | eta: 6.2m step 05256/05568 (94.40%) | loss: 2.376471 | lrm: 0.13 | dt: 1191.19ms | tok/sec: 880,274 | bf16_mfu: 54.07 | total time: 104.55m | eta: 6.2m step 05257/05568 (94.41%) | loss: 2.376754 | lrm: 0.13 | dt: 1192.84ms | tok/sec: 879,055 | bf16_mfu: 54.00 | total time: 104.57m | eta: 6.2m step 05258/05568 (94.43%) | loss: 2.379561 | lrm: 0.13 | dt: 1192.38ms | tok/sec: 879,399 | bf16_mfu: 54.02 | total time: 104.58m | eta: 6.2m step 05259/05568 (94.45%) | loss: 2.369504 | lrm: 0.13 | dt: 1193.05ms | tok/sec: 878,904 | bf16_mfu: 53.99 | total time: 104.60m | eta: 6.1m step 05260/05568 (94.47%) | loss: 2.380844 | lrm: 0.13 | dt: 1201.67ms | tok/sec: 872,596 | bf16_mfu: 53.60 | total time: 104.62m | eta: 6.1m step 05261/05568 (94.49%) | loss: 2.382529 | lrm: 0.13 | dt: 1190.97ms | tok/sec: 880,440 | bf16_mfu: 54.08 | total time: 104.64m | eta: 6.1m step 05262/05568 (94.50%) | loss: 2.374604 | lrm: 0.13 | dt: 1196.05ms | tok/sec: 876,698 | bf16_mfu: 53.85 | total time: 104.66m | eta: 6.1m step 05263/05568 (94.52%) | loss: 2.372742 | lrm: 0.13 | dt: 1195.86ms | tok/sec: 876,839 | bf16_mfu: 53.86 | total time: 104.68m | eta: 6.1m step 05264/05568 (94.54%) | loss: 2.380715 | lrm: 0.13 | dt: 1193.63ms | tok/sec: 878,474 | bf16_mfu: 53.96 | total time: 104.70m | eta: 6.0m step 05265/05568 (94.56%) | loss: 2.379818 | lrm: 0.13 | dt: 1194.59ms | tok/sec: 877,773 | bf16_mfu: 53.92 | total time: 104.72m | eta: 6.0m step 05266/05568 (94.58%) | loss: 2.384489 | lrm: 0.13 | dt: 1194.42ms | tok/sec: 877,893 | bf16_mfu: 53.93 | total time: 104.74m | eta: 6.0m step 05267/05568 (94.59%) | loss: 2.378934 | lrm: 0.13 | dt: 1196.51ms | tok/sec: 876,362 | bf16_mfu: 53.83 | total time: 104.76m | eta: 6.0m step 05268/05568 (94.61%) | loss: 2.378934 | lrm: 0.13 | dt: 1188.84ms | tok/sec: 882,017 | bf16_mfu: 54.18 | total time: 104.78m | eta: 6.0m step 05269/05568 (94.63%) | loss: 2.367467 | lrm: 0.13 | dt: 1193.18ms | tok/sec: 878,808 | bf16_mfu: 53.98 | total time: 104.80m | eta: 5.9m step 05270/05568 (94.65%) | loss: 2.369803 | lrm: 0.13 | dt: 1191.28ms | tok/sec: 880,213 | bf16_mfu: 54.07 | total time: 104.82m | eta: 5.9m step 05271/05568 (94.67%) | loss: 2.371334 | lrm: 0.13 | dt: 1198.40ms | tok/sec: 874,982 | bf16_mfu: 53.75 | total time: 104.84m | eta: 5.9m step 05272/05568 (94.68%) | loss: 2.368858 | lrm: 0.13 | dt: 1193.08ms | tok/sec: 878,883 | bf16_mfu: 53.99 | total time: 104.86m | eta: 5.9m step 05273/05568 (94.70%) | loss: 2.376860 | lrm: 0.13 | dt: 1191.22ms | tok/sec: 880,255 | bf16_mfu: 54.07 | total time: 104.88m | eta: 5.9m step 05274/05568 (94.72%) | loss: 2.379094 | lrm: 0.13 | dt: 1190.53ms | tok/sec: 880,763 | bf16_mfu: 54.10 | total time: 104.90m | eta: 5.8m step 05275/05568 (94.74%) | loss: 2.384919 | lrm: 0.13 | dt: 1196.28ms | tok/sec: 876,528 | bf16_mfu: 53.84 | total time: 104.92m | eta: 5.8m step 05276/05568 (94.76%) | loss: 2.373719 | lrm: 0.13 | dt: 1197.05ms | tok/sec: 875,969 | bf16_mfu: 53.81 | total time: 104.94m | eta: 5.8m step 05277/05568 (94.77%) | loss: 2.360143 | lrm: 0.13 | dt: 1196.63ms | tok/sec: 876,277 | bf16_mfu: 53.83 | total time: 104.96m | eta: 5.8m step 05278/05568 (94.79%) | loss: 2.354743 | lrm: 0.13 | dt: 1193.58ms | tok/sec: 878,509 | bf16_mfu: 53.96 | total time: 104.98m | eta: 5.8m step 05279/05568 (94.81%) | loss: 2.353645 | lrm: 0.13 | dt: 1189.98ms | tok/sec: 881,168 | bf16_mfu: 54.13 | total time: 105.00m | eta: 5.7m step 05280/05568 (94.83%) | loss: 2.343955 | lrm: 0.13 | dt: 1192.77ms | tok/sec: 879,107 | bf16_mfu: 54.00 | total time: 105.02m | eta: 5.7m step 05281/05568 (94.85%) | loss: 2.350151 | lrm: 0.13 | dt: 1193.78ms | tok/sec: 878,366 | bf16_mfu: 53.96 | total time: 105.04m | eta: 5.7m step 05282/05568 (94.86%) | loss: 2.348010 | lrm: 0.13 | dt: 1195.58ms | tok/sec: 877,046 | bf16_mfu: 53.87 | total time: 105.06m | eta: 5.7m step 05283/05568 (94.88%) | loss: 2.355934 | lrm: 0.12 | dt: 1195.40ms | tok/sec: 877,176 | bf16_mfu: 53.88 | total time: 105.08m | eta: 5.7m step 05284/05568 (94.90%) | loss: 2.355850 | lrm: 0.12 | dt: 1190.86ms | tok/sec: 880,519 | bf16_mfu: 54.09 | total time: 105.10m | eta: 5.6m step 05285/05568 (94.92%) | loss: 2.355706 | lrm: 0.12 | dt: 1198.34ms | tok/sec: 875,021 | bf16_mfu: 53.75 | total time: 105.12m | eta: 5.6m step 05286/05568 (94.94%) | loss: 2.348649 | lrm: 0.12 | dt: 1193.23ms | tok/sec: 878,769 | bf16_mfu: 53.98 | total time: 105.14m | eta: 5.6m step 05287/05568 (94.95%) | loss: 2.346141 | lrm: 0.12 | dt: 1192.41ms | tok/sec: 879,378 | bf16_mfu: 54.02 | total time: 105.16m | eta: 5.6m step 05288/05568 (94.97%) | loss: 2.342329 | lrm: 0.12 | dt: 1191.12ms | tok/sec: 880,331 | bf16_mfu: 54.08 | total time: 105.18m | eta: 5.6m step 05289/05568 (94.99%) | loss: 2.347515 | lrm: 0.12 | dt: 1195.92ms | tok/sec: 876,794 | bf16_mfu: 53.86 | total time: 105.20m | eta: 5.5m step 05290/05568 (95.01%) | loss: 2.335788 | lrm: 0.12 | dt: 1192.35ms | tok/sec: 879,418 | bf16_mfu: 54.02 | total time: 105.22m | eta: 5.5m step 05291/05568 (95.03%) | loss: 2.333711 | lrm: 0.12 | dt: 1195.53ms | tok/sec: 877,079 | bf16_mfu: 53.88 | total time: 105.24m | eta: 5.5m step 05292/05568 (95.04%) | loss: 2.333736 | lrm: 0.12 | dt: 1192.39ms | tok/sec: 879,387 | bf16_mfu: 54.02 | total time: 105.26m | eta: 5.5m step 05293/05568 (95.06%) | loss: 2.340211 | lrm: 0.12 | dt: 1197.39ms | tok/sec: 875,717 | bf16_mfu: 53.79 | total time: 105.28m | eta: 5.5m step 05294/05568 (95.08%) | loss: 2.348211 | lrm: 0.12 | dt: 1187.38ms | tok/sec: 883,103 | bf16_mfu: 54.25 | total time: 105.30m | eta: 5.4m step 05295/05568 (95.10%) | loss: 2.348423 | lrm: 0.12 | dt: 1195.22ms | tok/sec: 877,306 | bf16_mfu: 53.89 | total time: 105.32m | eta: 5.4m step 05296/05568 (95.11%) | loss: 2.343666 | lrm: 0.12 | dt: 1197.18ms | tok/sec: 875,870 | bf16_mfu: 53.80 | total time: 105.34m | eta: 5.4m step 05297/05568 (95.13%) | loss: 2.358446 | lrm: 0.12 | dt: 1191.38ms | tok/sec: 880,133 | bf16_mfu: 54.06 | total time: 105.36m | eta: 5.4m step 05298/05568 (95.15%) | loss: 2.359100 | lrm: 0.12 | dt: 1192.15ms | tok/sec: 879,566 | bf16_mfu: 54.03 | total time: 105.38m | eta: 5.4m step 05299/05568 (95.17%) | loss: 2.368977 | lrm: 0.12 | dt: 1197.13ms | tok/sec: 875,909 | bf16_mfu: 53.81 | total time: 105.40m | eta: 5.3m step 05300/05568 (95.19%) | loss: 2.361826 | lrm: 0.12 | dt: 1190.41ms | tok/sec: 880,851 | bf16_mfu: 54.11 | total time: 105.42m | eta: 5.3m step 05301/05568 (95.20%) | loss: 2.370686 | lrm: 0.12 | dt: 1194.30ms | tok/sec: 877,980 | bf16_mfu: 53.93 | total time: 105.44m | eta: 5.3m step 05302/05568 (95.22%) | loss: 2.373718 | lrm: 0.12 | dt: 1192.17ms | tok/sec: 879,551 | bf16_mfu: 54.03 | total time: 105.46m | eta: 5.3m step 05303/05568 (95.24%) | loss: 2.366802 | lrm: 0.12 | dt: 1188.83ms | tok/sec: 882,025 | bf16_mfu: 54.18 | total time: 105.48m | eta: 5.3m step 05304/05568 (95.26%) | loss: 2.372572 | lrm: 0.12 | dt: 1193.97ms | tok/sec: 878,227 | bf16_mfu: 53.95 | total time: 105.50m | eta: 5.2m step 05305/05568 (95.28%) | loss: 2.368584 | lrm: 0.12 | dt: 1187.06ms | tok/sec: 883,340 | bf16_mfu: 54.26 | total time: 105.52m | eta: 5.2m step 05306/05568 (95.29%) | loss: 2.365266 | lrm: 0.12 | dt: 1195.96ms | tok/sec: 876,766 | bf16_mfu: 53.86 | total time: 105.54m | eta: 5.2m step 05307/05568 (95.31%) | loss: 2.358326 | lrm: 0.12 | dt: 1196.42ms | tok/sec: 876,430 | bf16_mfu: 53.84 | total time: 105.56m | eta: 5.2m step 05308/05568 (95.33%) | loss: 2.359156 | lrm: 0.12 | dt: 1196.00ms | tok/sec: 876,732 | bf16_mfu: 53.86 | total time: 105.58m | eta: 5.2m step 05309/05568 (95.35%) | loss: 2.358340 | lrm: 0.12 | dt: 1195.38ms | tok/sec: 877,192 | bf16_mfu: 53.88 | total time: 105.60m | eta: 5.1m step 05310/05568 (95.37%) | loss: 2.361292 | lrm: 0.12 | dt: 1192.29ms | tok/sec: 879,460 | bf16_mfu: 54.02 | total time: 105.62m | eta: 5.1m step 05311/05568 (95.38%) | loss: 2.366900 | lrm: 0.12 | dt: 1195.45ms | tok/sec: 877,139 | bf16_mfu: 53.88 | total time: 105.64m | eta: 5.1m step 05312/05568 (95.40%) | loss: 2.362800 | lrm: 0.12 | dt: 1201.93ms | tok/sec: 872,410 | bf16_mfu: 53.59 | total time: 105.66m | eta: 5.1m step 05313/05568 (95.42%) | loss: 2.378309 | lrm: 0.12 | dt: 1197.64ms | tok/sec: 875,533 | bf16_mfu: 53.78 | total time: 105.68m | eta: 5.1m step 05314/05568 (95.44%) | loss: 2.381775 | lrm: 0.12 | dt: 1190.91ms | tok/sec: 880,486 | bf16_mfu: 54.09 | total time: 105.70m | eta: 5.0m step 05315/05568 (95.46%) | loss: 2.365895 | lrm: 0.12 | dt: 1194.23ms | tok/sec: 878,038 | bf16_mfu: 53.94 | total time: 105.72m | eta: 5.0m step 05316/05568 (95.47%) | loss: 2.373218 | lrm: 0.12 | dt: 1190.39ms | tok/sec: 880,869 | bf16_mfu: 54.11 | total time: 105.74m | eta: 5.0m step 05317/05568 (95.49%) | loss: 2.360587 | lrm: 0.12 | dt: 1193.39ms | tok/sec: 878,650 | bf16_mfu: 53.97 | total time: 105.76m | eta: 5.0m step 05318/05568 (95.51%) | loss: 2.358820 | lrm: 0.12 | dt: 1191.57ms | tok/sec: 879,993 | bf16_mfu: 54.06 | total time: 105.78m | eta: 5.0m step 05319/05568 (95.53%) | loss: 2.367172 | lrm: 0.12 | dt: 1189.96ms | tok/sec: 881,187 | bf16_mfu: 54.13 | total time: 105.80m | eta: 4.9m step 05320/05568 (95.55%) | loss: 2.355293 | lrm: 0.12 | dt: 1194.05ms | tok/sec: 878,167 | bf16_mfu: 53.94 | total time: 105.82m | eta: 4.9m step 05321/05568 (95.56%) | loss: 2.373150 | lrm: 0.11 | dt: 1197.06ms | tok/sec: 875,963 | bf16_mfu: 53.81 | total time: 105.84m | eta: 4.9m step 05322/05568 (95.58%) | loss: 2.370206 | lrm: 0.11 | dt: 1192.25ms | tok/sec: 879,492 | bf16_mfu: 54.03 | total time: 105.86m | eta: 4.9m step 05323/05568 (95.60%) | loss: 2.360095 | lrm: 0.11 | dt: 1189.65ms | tok/sec: 881,417 | bf16_mfu: 54.14 | total time: 105.88m | eta: 4.9m step 05324/05568 (95.62%) | loss: 2.361731 | lrm: 0.11 | dt: 1197.12ms | tok/sec: 875,917 | bf16_mfu: 53.81 | total time: 105.90m | eta: 4.8m step 05325/05568 (95.64%) | loss: 2.369253 | lrm: 0.11 | dt: 1198.00ms | tok/sec: 875,271 | bf16_mfu: 53.77 | total time: 105.92m | eta: 4.8m step 05326/05568 (95.65%) | loss: 2.368449 | lrm: 0.11 | dt: 1197.96ms | tok/sec: 875,302 | bf16_mfu: 53.77 | total time: 105.94m | eta: 4.8m step 05327/05568 (95.67%) | loss: 2.373339 | lrm: 0.11 | dt: 1193.09ms | tok/sec: 878,874 | bf16_mfu: 53.99 | total time: 105.96m | eta: 4.8m step 05328/05568 (95.69%) | loss: 2.370814 | lrm: 0.11 | dt: 1196.91ms | tok/sec: 876,071 | bf16_mfu: 53.82 | total time: 105.98m | eta: 4.8m step 05329/05568 (95.71%) | loss: 2.373391 | lrm: 0.11 | dt: 1197.52ms | tok/sec: 875,621 | bf16_mfu: 53.79 | total time: 106.00m | eta: 4.7m step 05330/05568 (95.73%) | loss: 2.376865 | lrm: 0.11 | dt: 1193.55ms | tok/sec: 878,533 | bf16_mfu: 53.97 | total time: 106.02m | eta: 4.7m step 05331/05568 (95.74%) | loss: 2.358214 | lrm: 0.11 | dt: 1191.29ms | tok/sec: 880,198 | bf16_mfu: 54.07 | total time: 106.04m | eta: 4.7m step 05332/05568 (95.76%) | loss: 2.354066 | lrm: 0.11 | dt: 1195.85ms | tok/sec: 876,844 | bf16_mfu: 53.86 | total time: 106.06m | eta: 4.7m step 05333/05568 (95.78%) | loss: 2.360487 | lrm: 0.11 | dt: 1196.55ms | tok/sec: 876,334 | bf16_mfu: 53.83 | total time: 106.08m | eta: 4.7m step 05334/05568 (95.80%) | loss: 2.359935 | lrm: 0.11 | dt: 1195.36ms | tok/sec: 877,203 | bf16_mfu: 53.88 | total time: 106.10m | eta: 4.6m step 05335/05568 (95.82%) | loss: 2.367878 | lrm: 0.11 | dt: 1193.95ms | tok/sec: 878,240 | bf16_mfu: 53.95 | total time: 106.12m | eta: 4.6m step 05336/05568 (95.83%) | loss: 2.380088 | lrm: 0.11 | dt: 1195.08ms | tok/sec: 877,407 | bf16_mfu: 53.90 | total time: 106.14m | eta: 4.6m step 05337/05568 (95.85%) | loss: 2.378282 | lrm: 0.11 | dt: 1195.62ms | tok/sec: 877,013 | bf16_mfu: 53.87 | total time: 106.16m | eta: 4.6m step 05338/05568 (95.87%) | loss: 2.382242 | lrm: 0.11 | dt: 1194.09ms | tok/sec: 878,140 | bf16_mfu: 53.94 | total time: 106.18m | eta: 4.6m step 05339/05568 (95.89%) | loss: 2.390675 | lrm: 0.11 | dt: 1198.05ms | tok/sec: 875,234 | bf16_mfu: 53.76 | total time: 106.20m | eta: 4.5m step 05340/05568 (95.91%) | loss: 2.393586 | lrm: 0.11 | dt: 1194.78ms | tok/sec: 877,634 | bf16_mfu: 53.91 | total time: 106.22m | eta: 4.5m step 05341/05568 (95.92%) | loss: 2.393423 | lrm: 0.11 | dt: 1196.19ms | tok/sec: 876,596 | bf16_mfu: 53.85 | total time: 106.24m | eta: 4.5m step 05342/05568 (95.94%) | loss: 2.386022 | lrm: 0.11 | dt: 1197.43ms | tok/sec: 875,687 | bf16_mfu: 53.79 | total time: 106.26m | eta: 4.5m step 05343/05568 (95.96%) | loss: 2.386765 | lrm: 0.11 | dt: 1196.72ms | tok/sec: 876,210 | bf16_mfu: 53.82 | total time: 106.28m | eta: 4.5m step 05344/05568 (95.98%) | loss: 2.383572 | lrm: 0.11 | dt: 1196.00ms | tok/sec: 876,739 | bf16_mfu: 53.86 | total time: 106.30m | eta: 4.4m step 05345/05568 (95.99%) | loss: 2.379478 | lrm: 0.11 | dt: 1194.26ms | tok/sec: 878,016 | bf16_mfu: 53.93 | total time: 106.32m | eta: 4.4m step 05346/05568 (96.01%) | loss: 2.375331 | lrm: 0.11 | dt: 1192.83ms | tok/sec: 879,068 | bf16_mfu: 54.00 | total time: 106.34m | eta: 4.4m step 05347/05568 (96.03%) | loss: 2.370706 | lrm: 0.11 | dt: 1195.26ms | tok/sec: 877,281 | bf16_mfu: 53.89 | total time: 106.36m | eta: 4.4m step 05348/05568 (96.05%) | loss: 2.365502 | lrm: 0.11 | dt: 1197.55ms | tok/sec: 875,602 | bf16_mfu: 53.79 | total time: 106.38m | eta: 4.4m step 05349/05568 (96.07%) | loss: 2.369726 | lrm: 0.11 | dt: 1192.26ms | tok/sec: 879,486 | bf16_mfu: 54.02 | total time: 106.40m | eta: 4.3m step 05350/05568 (96.08%) | loss: 2.364959 | lrm: 0.11 | dt: 1198.92ms | tok/sec: 874,599 | bf16_mfu: 53.72 | total time: 106.42m | eta: 4.3m step 05351/05568 (96.10%) | loss: 2.365957 | lrm: 0.11 | dt: 1195.10ms | tok/sec: 877,397 | bf16_mfu: 53.90 | total time: 106.44m | eta: 4.3m step 05352/05568 (96.12%) | loss: 2.365064 | lrm: 0.11 | dt: 1195.73ms | tok/sec: 876,935 | bf16_mfu: 53.87 | total time: 106.46m | eta: 4.3m step 05353/05568 (96.14%) | loss: 2.352104 | lrm: 0.11 | dt: 1194.13ms | tok/sec: 878,111 | bf16_mfu: 53.94 | total time: 106.48m | eta: 4.3m step 05354/05568 (96.16%) | loss: 2.347208 | lrm: 0.11 | dt: 1194.43ms | tok/sec: 877,886 | bf16_mfu: 53.93 | total time: 106.50m | eta: 4.2m step 05355/05568 (96.17%) | loss: 2.354569 | lrm: 0.11 | dt: 1193.24ms | tok/sec: 878,766 | bf16_mfu: 53.98 | total time: 106.52m | eta: 4.2m step 05356/05568 (96.19%) | loss: 2.349728 | lrm: 0.11 | dt: 1195.65ms | tok/sec: 876,995 | bf16_mfu: 53.87 | total time: 106.54m | eta: 4.2m step 05357/05568 (96.21%) | loss: 2.348992 | lrm: 0.11 | dt: 1196.99ms | tok/sec: 876,014 | bf16_mfu: 53.81 | total time: 106.56m | eta: 4.2m step 05358/05568 (96.23%) | loss: 2.352046 | lrm: 0.11 | dt: 1194.19ms | tok/sec: 878,063 | bf16_mfu: 53.94 | total time: 106.58m | eta: 4.2m step 05359/05568 (96.25%) | loss: 2.352018 | lrm: 0.10 | dt: 1193.54ms | tok/sec: 878,546 | bf16_mfu: 53.97 | total time: 106.60m | eta: 4.1m step 05360/05568 (96.26%) | loss: 2.359958 | lrm: 0.10 | dt: 1194.77ms | tok/sec: 877,637 | bf16_mfu: 53.91 | total time: 106.62m | eta: 4.1m step 05361/05568 (96.28%) | loss: 2.367475 | lrm: 0.10 | dt: 1196.73ms | tok/sec: 876,201 | bf16_mfu: 53.82 | total time: 106.64m | eta: 4.1m step 05362/05568 (96.30%) | loss: 2.365871 | lrm: 0.10 | dt: 1198.46ms | tok/sec: 874,933 | bf16_mfu: 53.75 | total time: 106.66m | eta: 4.1m step 05363/05568 (96.32%) | loss: 2.361726 | lrm: 0.10 | dt: 1197.13ms | tok/sec: 875,908 | bf16_mfu: 53.81 | total time: 106.68m | eta: 4.1m step 05364/05568 (96.34%) | loss: 2.355796 | lrm: 0.10 | dt: 1190.01ms | tok/sec: 881,146 | bf16_mfu: 54.13 | total time: 106.70m | eta: 4.0m step 05365/05568 (96.35%) | loss: 2.360908 | lrm: 0.10 | dt: 1191.76ms | tok/sec: 879,855 | bf16_mfu: 54.05 | total time: 106.71m | eta: 4.0m step 05366/05568 (96.37%) | loss: 2.357442 | lrm: 0.10 | dt: 1198.15ms | tok/sec: 875,164 | bf16_mfu: 53.76 | total time: 106.73m | eta: 4.0m step 05367/05568 (96.39%) | loss: 2.366752 | lrm: 0.10 | dt: 1192.40ms | tok/sec: 879,384 | bf16_mfu: 54.02 | total time: 106.75m | eta: 4.0m step 05368/05568 (96.41%) | loss: 2.372975 | lrm: 0.10 | dt: 1198.19ms | tok/sec: 875,132 | bf16_mfu: 53.76 | total time: 106.77m | eta: 4.0m step 05369/05568 (96.43%) | loss: 2.376144 | lrm: 0.10 | dt: 1194.71ms | tok/sec: 877,680 | bf16_mfu: 53.91 | total time: 106.79m | eta: 3.9m step 05370/05568 (96.44%) | loss: 2.380976 | lrm: 0.10 | dt: 1194.21ms | tok/sec: 878,049 | bf16_mfu: 53.94 | total time: 106.81m | eta: 3.9m step 05371/05568 (96.46%) | loss: 2.382304 | lrm: 0.10 | dt: 1193.79ms | tok/sec: 878,362 | bf16_mfu: 53.96 | total time: 106.83m | eta: 3.9m step 05372/05568 (96.48%) | loss: 2.379628 | lrm: 0.10 | dt: 1194.86ms | tok/sec: 877,569 | bf16_mfu: 53.91 | total time: 106.85m | eta: 3.9m step 05373/05568 (96.50%) | loss: 2.369782 | lrm: 0.10 | dt: 1194.90ms | tok/sec: 877,541 | bf16_mfu: 53.91 | total time: 106.87m | eta: 3.9m step 05374/05568 (96.52%) | loss: 2.360026 | lrm: 0.10 | dt: 1191.25ms | tok/sec: 880,228 | bf16_mfu: 54.07 | total time: 106.89m | eta: 3.8m step 05375/05568 (96.53%) | loss: 2.371274 | lrm: 0.10 | dt: 1194.65ms | tok/sec: 877,724 | bf16_mfu: 53.92 | total time: 106.91m | eta: 3.8m step 05376/05568 (96.55%) | loss: 2.372160 | lrm: 0.10 | dt: 1198.05ms | tok/sec: 875,233 | bf16_mfu: 53.76 | total time: 106.93m | eta: 3.8m step 05377/05568 (96.57%) | loss: 2.367947 | lrm: 0.10 | dt: 1196.67ms | tok/sec: 876,244 | bf16_mfu: 53.83 | total time: 106.95m | eta: 3.8m step 05378/05568 (96.59%) | loss: 2.366827 | lrm: 0.10 | dt: 1187.94ms | tok/sec: 882,686 | bf16_mfu: 54.22 | total time: 106.97m | eta: 3.8m step 05379/05568 (96.61%) | loss: 2.364367 | lrm: 0.10 | dt: 1192.62ms | tok/sec: 879,222 | bf16_mfu: 54.01 | total time: 106.99m | eta: 3.7m step 05380/05568 (96.62%) | loss: 2.355699 | lrm: 0.10 | dt: 1194.30ms | tok/sec: 877,987 | bf16_mfu: 53.93 | total time: 107.01m | eta: 3.7m step 05381/05568 (96.64%) | loss: 2.353528 | lrm: 0.10 | dt: 1192.43ms | tok/sec: 879,362 | bf16_mfu: 54.02 | total time: 107.03m | eta: 3.7m step 05382/05568 (96.66%) | loss: 2.352385 | lrm: 0.10 | dt: 1194.62ms | tok/sec: 877,746 | bf16_mfu: 53.92 | total time: 107.05m | eta: 3.7m step 05383/05568 (96.68%) | loss: 2.362937 | lrm: 0.10 | dt: 1197.01ms | tok/sec: 875,998 | bf16_mfu: 53.81 | total time: 107.07m | eta: 3.7m step 05384/05568 (96.70%) | loss: 2.347531 | lrm: 0.10 | dt: 1193.36ms | tok/sec: 878,672 | bf16_mfu: 53.97 | total time: 107.09m | eta: 3.6m step 05385/05568 (96.71%) | loss: 2.338885 | lrm: 0.10 | dt: 1193.49ms | tok/sec: 878,579 | bf16_mfu: 53.97 | total time: 107.11m | eta: 3.6m step 05386/05568 (96.73%) | loss: 2.351349 | lrm: 0.10 | dt: 1198.16ms | tok/sec: 875,156 | bf16_mfu: 53.76 | total time: 107.13m | eta: 3.6m step 05387/05568 (96.75%) | loss: 2.352473 | lrm: 0.10 | dt: 1195.68ms | tok/sec: 876,969 | bf16_mfu: 53.87 | total time: 107.15m | eta: 3.6m step 05388/05568 (96.77%) | loss: 2.355517 | lrm: 0.10 | dt: 1199.99ms | tok/sec: 873,819 | bf16_mfu: 53.68 | total time: 107.17m | eta: 3.6m step 05389/05568 (96.79%) | loss: 2.351616 | lrm: 0.10 | dt: 1189.02ms | tok/sec: 881,879 | bf16_mfu: 54.17 | total time: 107.19m | eta: 3.5m step 05390/05568 (96.80%) | loss: 2.346491 | lrm: 0.10 | dt: 1197.47ms | tok/sec: 875,662 | bf16_mfu: 53.79 | total time: 107.21m | eta: 3.5m step 05391/05568 (96.82%) | loss: 2.354315 | lrm: 0.10 | dt: 1192.70ms | tok/sec: 879,163 | bf16_mfu: 54.01 | total time: 107.23m | eta: 3.5m step 05392/05568 (96.84%) | loss: 2.357337 | lrm: 0.10 | dt: 1196.64ms | tok/sec: 876,268 | bf16_mfu: 53.83 | total time: 107.25m | eta: 3.5m step 05393/05568 (96.86%) | loss: 2.363849 | lrm: 0.10 | dt: 1194.96ms | tok/sec: 877,495 | bf16_mfu: 53.90 | total time: 107.27m | eta: 3.5m step 05394/05568 (96.88%) | loss: 2.369559 | lrm: 0.10 | dt: 1196.57ms | tok/sec: 876,316 | bf16_mfu: 53.83 | total time: 107.29m | eta: 3.4m step 05395/05568 (96.89%) | loss: 2.355361 | lrm: 0.10 | dt: 1193.43ms | tok/sec: 878,620 | bf16_mfu: 53.97 | total time: 107.31m | eta: 3.4m step 05396/05568 (96.91%) | loss: 2.357313 | lrm: 0.10 | dt: 1195.23ms | tok/sec: 877,302 | bf16_mfu: 53.89 | total time: 107.33m | eta: 3.4m step 05397/05568 (96.93%) | loss: 2.342657 | lrm: 0.09 | dt: 1191.57ms | tok/sec: 879,993 | bf16_mfu: 54.06 | total time: 107.35m | eta: 3.4m step 05398/05568 (96.95%) | loss: 2.344622 | lrm: 0.09 | dt: 1191.64ms | tok/sec: 879,942 | bf16_mfu: 54.05 | total time: 107.37m | eta: 3.4m step 05399/05568 (96.96%) | loss: 2.349892 | lrm: 0.09 | dt: 1194.82ms | tok/sec: 877,599 | bf16_mfu: 53.91 | total time: 107.39m | eta: 3.3m step 05400/05568 (96.98%) | loss: 2.345329 | lrm: 0.09 | dt: 1194.85ms | tok/sec: 877,583 | bf16_mfu: 53.91 | total time: 107.41m | eta: 3.3m step 05401/05568 (97.00%) | loss: 2.344876 | lrm: 0.09 | dt: 1197.28ms | tok/sec: 875,800 | bf16_mfu: 53.80 | total time: 107.43m | eta: 3.3m step 05402/05568 (97.02%) | loss: 2.345437 | lrm: 0.09 | dt: 1195.53ms | tok/sec: 877,078 | bf16_mfu: 53.88 | total time: 107.45m | eta: 3.3m step 05403/05568 (97.04%) | loss: 2.346928 | lrm: 0.09 | dt: 1194.82ms | tok/sec: 877,601 | bf16_mfu: 53.91 | total time: 107.47m | eta: 3.3m step 05404/05568 (97.05%) | loss: 2.353648 | lrm: 0.09 | dt: 1195.05ms | tok/sec: 877,429 | bf16_mfu: 53.90 | total time: 107.49m | eta: 3.2m step 05405/05568 (97.07%) | loss: 2.345895 | lrm: 0.09 | dt: 1193.00ms | tok/sec: 878,936 | bf16_mfu: 53.99 | total time: 107.51m | eta: 3.2m step 05406/05568 (97.09%) | loss: 2.343447 | lrm: 0.09 | dt: 1196.43ms | tok/sec: 876,421 | bf16_mfu: 53.84 | total time: 107.53m | eta: 3.2m step 05407/05568 (97.11%) | loss: 2.336881 | lrm: 0.09 | dt: 1192.47ms | tok/sec: 879,331 | bf16_mfu: 54.02 | total time: 107.55m | eta: 3.2m step 05408/05568 (97.13%) | loss: 2.337990 | lrm: 0.09 | dt: 1194.18ms | tok/sec: 878,075 | bf16_mfu: 53.94 | total time: 107.57m | eta: 3.2m step 05409/05568 (97.14%) | loss: 2.349258 | lrm: 0.09 | dt: 1192.16ms | tok/sec: 879,563 | bf16_mfu: 54.03 | total time: 107.59m | eta: 3.1m step 05410/05568 (97.16%) | loss: 2.339867 | lrm: 0.09 | dt: 1197.03ms | tok/sec: 875,981 | bf16_mfu: 53.81 | total time: 107.61m | eta: 3.1m step 05411/05568 (97.18%) | loss: 2.344881 | lrm: 0.09 | dt: 1197.32ms | tok/sec: 875,772 | bf16_mfu: 53.80 | total time: 107.63m | eta: 3.1m step 05412/05568 (97.20%) | loss: 2.352054 | lrm: 0.09 | dt: 1194.91ms | tok/sec: 877,533 | bf16_mfu: 53.90 | total time: 107.65m | eta: 3.1m step 05413/05568 (97.22%) | loss: 2.367796 | lrm: 0.09 | dt: 1194.33ms | tok/sec: 877,965 | bf16_mfu: 53.93 | total time: 107.67m | eta: 3.1m step 05414/05568 (97.23%) | loss: 2.374237 | lrm: 0.09 | dt: 1195.47ms | tok/sec: 877,122 | bf16_mfu: 53.88 | total time: 107.69m | eta: 3.0m step 05415/05568 (97.25%) | loss: 2.380828 | lrm: 0.09 | dt: 1192.76ms | tok/sec: 879,116 | bf16_mfu: 54.00 | total time: 107.71m | eta: 3.0m step 05416/05568 (97.27%) | loss: 2.379142 | lrm: 0.09 | dt: 1197.98ms | tok/sec: 875,287 | bf16_mfu: 53.77 | total time: 107.73m | eta: 3.0m step 05417/05568 (97.29%) | loss: 2.374722 | lrm: 0.09 | dt: 1192.97ms | tok/sec: 878,961 | bf16_mfu: 53.99 | total time: 107.75m | eta: 3.0m step 05418/05568 (97.31%) | loss: 2.379549 | lrm: 0.09 | dt: 1194.20ms | tok/sec: 878,056 | bf16_mfu: 53.94 | total time: 107.77m | eta: 3.0m step 05419/05568 (97.32%) | loss: 2.376729 | lrm: 0.09 | dt: 1197.01ms | tok/sec: 875,996 | bf16_mfu: 53.81 | total time: 107.79m | eta: 2.9m step 05420/05568 (97.34%) | loss: 2.386411 | lrm: 0.09 | dt: 1196.86ms | tok/sec: 876,103 | bf16_mfu: 53.82 | total time: 107.81m | eta: 2.9m step 05421/05568 (97.36%) | loss: 2.385411 | lrm: 0.09 | dt: 1193.37ms | tok/sec: 878,669 | bf16_mfu: 53.97 | total time: 107.83m | eta: 2.9m step 05422/05568 (97.38%) | loss: 2.394865 | lrm: 0.09 | dt: 1196.13ms | tok/sec: 876,642 | bf16_mfu: 53.85 | total time: 107.85m | eta: 2.9m step 05423/05568 (97.40%) | loss: 2.398057 | lrm: 0.09 | dt: 1197.46ms | tok/sec: 875,664 | bf16_mfu: 53.79 | total time: 107.87m | eta: 2.9m step 05424/05568 (97.41%) | loss: 2.399803 | lrm: 0.09 | dt: 1196.93ms | tok/sec: 876,051 | bf16_mfu: 53.81 | total time: 107.89m | eta: 2.8m step 05425/05568 (97.43%) | loss: 2.396894 | lrm: 0.09 | dt: 1194.71ms | tok/sec: 877,679 | bf16_mfu: 53.91 | total time: 107.91m | eta: 2.8m step 05426/05568 (97.45%) | loss: 2.395995 | lrm: 0.09 | dt: 1196.07ms | tok/sec: 876,684 | bf16_mfu: 53.85 | total time: 107.93m | eta: 2.8m step 05427/05568 (97.47%) | loss: 2.392511 | lrm: 0.09 | dt: 1193.02ms | tok/sec: 878,927 | bf16_mfu: 53.99 | total time: 107.95m | eta: 2.8m step 05428/05568 (97.49%) | loss: 2.390398 | lrm: 0.09 | dt: 1197.43ms | tok/sec: 875,691 | bf16_mfu: 53.79 | total time: 107.97m | eta: 2.8m step 05429/05568 (97.50%) | loss: 2.387745 | lrm: 0.09 | dt: 1198.10ms | tok/sec: 875,195 | bf16_mfu: 53.76 | total time: 107.99m | eta: 2.8m step 05430/05568 (97.52%) | loss: 2.387270 | lrm: 0.09 | dt: 1197.75ms | tok/sec: 875,456 | bf16_mfu: 53.78 | total time: 108.01m | eta: 2.7m step 05431/05568 (97.54%) | loss: 2.396974 | lrm: 0.09 | dt: 1197.82ms | tok/sec: 875,402 | bf16_mfu: 53.77 | total time: 108.03m | eta: 2.7m step 05432/05568 (97.56%) | loss: 2.403933 | lrm: 0.09 | dt: 1194.19ms | tok/sec: 878,067 | bf16_mfu: 53.94 | total time: 108.05m | eta: 2.7m step 05433/05568 (97.58%) | loss: 2.390860 | lrm: 0.09 | dt: 1197.02ms | tok/sec: 875,990 | bf16_mfu: 53.81 | total time: 108.07m | eta: 2.7m step 05434/05568 (97.59%) | loss: 2.386356 | lrm: 0.09 | dt: 1193.59ms | tok/sec: 878,509 | bf16_mfu: 53.96 | total time: 108.09m | eta: 2.7m step 05435/05568 (97.61%) | loss: 2.374105 | lrm: 0.08 | dt: 1197.44ms | tok/sec: 875,681 | bf16_mfu: 53.79 | total time: 108.11m | eta: 2.6m step 05436/05568 (97.63%) | loss: 2.383651 | lrm: 0.08 | dt: 1196.07ms | tok/sec: 876,686 | bf16_mfu: 53.85 | total time: 108.13m | eta: 2.6m step 05437/05568 (97.65%) | loss: 2.388686 | lrm: 0.08 | dt: 1194.35ms | tok/sec: 877,945 | bf16_mfu: 53.93 | total time: 108.15m | eta: 2.6m step 05438/05568 (97.67%) | loss: 2.389306 | lrm: 0.08 | dt: 1194.50ms | tok/sec: 877,837 | bf16_mfu: 53.92 | total time: 108.17m | eta: 2.6m step 05439/05568 (97.68%) | loss: 2.392387 | lrm: 0.08 | dt: 1194.59ms | tok/sec: 877,769 | bf16_mfu: 53.92 | total time: 108.19m | eta: 2.6m step 05440/05568 (97.70%) | loss: 2.387385 | lrm: 0.08 | dt: 1196.61ms | tok/sec: 876,286 | bf16_mfu: 53.83 | total time: 108.21m | eta: 2.5m step 05441/05568 (97.72%) | loss: 2.387660 | lrm: 0.08 | dt: 1192.91ms | tok/sec: 879,005 | bf16_mfu: 54.00 | total time: 108.23m | eta: 2.5m step 05442/05568 (97.74%) | loss: 2.382033 | lrm: 0.08 | dt: 1194.25ms | tok/sec: 878,023 | bf16_mfu: 53.93 | total time: 108.25m | eta: 2.5m step 05443/05568 (97.76%) | loss: 2.386570 | lrm: 0.08 | dt: 1193.75ms | tok/sec: 878,391 | bf16_mfu: 53.96 | total time: 108.27m | eta: 2.5m step 05444/05568 (97.77%) | loss: 2.399713 | lrm: 0.08 | dt: 1195.62ms | tok/sec: 877,014 | bf16_mfu: 53.87 | total time: 108.29m | eta: 2.5m step 05445/05568 (97.79%) | loss: 2.388833 | lrm: 0.08 | dt: 1196.51ms | tok/sec: 876,365 | bf16_mfu: 53.83 | total time: 108.31m | eta: 2.4m step 05446/05568 (97.81%) | loss: 2.395621 | lrm: 0.08 | dt: 1192.45ms | tok/sec: 879,343 | bf16_mfu: 54.02 | total time: 108.33m | eta: 2.4m step 05447/05568 (97.83%) | loss: 2.396596 | lrm: 0.08 | dt: 1197.11ms | tok/sec: 875,922 | bf16_mfu: 53.81 | total time: 108.35m | eta: 2.4m step 05448/05568 (97.84%) | loss: 2.401677 | lrm: 0.08 | dt: 1195.48ms | tok/sec: 877,117 | bf16_mfu: 53.88 | total time: 108.37m | eta: 2.4m step 05449/05568 (97.86%) | loss: 2.408255 | lrm: 0.08 | dt: 1193.75ms | tok/sec: 878,389 | bf16_mfu: 53.96 | total time: 108.39m | eta: 2.4m step 05450/05568 (97.88%) | loss: 2.399943 | lrm: 0.08 | dt: 1193.97ms | tok/sec: 878,229 | bf16_mfu: 53.95 | total time: 108.41m | eta: 2.3m step 05451/05568 (97.90%) | loss: 2.390816 | lrm: 0.08 | dt: 1196.95ms | tok/sec: 876,039 | bf16_mfu: 53.81 | total time: 108.43m | eta: 2.3m step 05452/05568 (97.92%) | loss: 2.390604 | lrm: 0.08 | dt: 1200.07ms | tok/sec: 873,765 | bf16_mfu: 53.67 | total time: 108.45m | eta: 2.3m step 05453/05568 (97.93%) | loss: 2.384234 | lrm: 0.08 | dt: 1192.18ms | tok/sec: 879,544 | bf16_mfu: 54.03 | total time: 108.47m | eta: 2.3m step 05454/05568 (97.95%) | loss: 2.384767 | lrm: 0.08 | dt: 1196.26ms | tok/sec: 876,546 | bf16_mfu: 53.84 | total time: 108.49m | eta: 2.3m step 05455/05568 (97.97%) | loss: 2.382723 | lrm: 0.08 | dt: 1192.33ms | tok/sec: 879,431 | bf16_mfu: 54.02 | total time: 108.51m | eta: 2.2m step 05456/05568 (97.99%) | loss: 2.378310 | lrm: 0.08 | dt: 1195.28ms | tok/sec: 877,262 | bf16_mfu: 53.89 | total time: 108.53m | eta: 2.2m step 05457/05568 (98.01%) | loss: 2.379802 | lrm: 0.08 | dt: 1196.00ms | tok/sec: 876,734 | bf16_mfu: 53.86 | total time: 108.55m | eta: 2.2m step 05458/05568 (98.02%) | loss: 2.380211 | lrm: 0.08 | dt: 1189.59ms | tok/sec: 881,459 | bf16_mfu: 54.15 | total time: 108.57m | eta: 2.2m step 05459/05568 (98.04%) | loss: 2.378073 | lrm: 0.08 | dt: 1197.18ms | tok/sec: 875,872 | bf16_mfu: 53.80 | total time: 108.59m | eta: 2.2m step 05460/05568 (98.06%) | loss: 2.380712 | lrm: 0.08 | dt: 1194.38ms | tok/sec: 877,928 | bf16_mfu: 53.93 | total time: 108.61m | eta: 2.1m step 05461/05568 (98.08%) | loss: 2.382869 | lrm: 0.08 | dt: 1193.26ms | tok/sec: 878,748 | bf16_mfu: 53.98 | total time: 108.63m | eta: 2.1m step 05462/05568 (98.10%) | loss: 2.376691 | lrm: 0.08 | dt: 1192.86ms | tok/sec: 879,045 | bf16_mfu: 54.00 | total time: 108.65m | eta: 2.1m step 05463/05568 (98.11%) | loss: 2.374369 | lrm: 0.08 | dt: 1196.86ms | tok/sec: 876,107 | bf16_mfu: 53.82 | total time: 108.67m | eta: 2.1m step 05464/05568 (98.13%) | loss: 2.382793 | lrm: 0.08 | dt: 1197.72ms | tok/sec: 875,476 | bf16_mfu: 53.78 | total time: 108.69m | eta: 2.1m step 05465/05568 (98.15%) | loss: 2.383949 | lrm: 0.08 | dt: 1197.11ms | tok/sec: 875,920 | bf16_mfu: 53.81 | total time: 108.71m | eta: 2.0m step 05466/05568 (98.17%) | loss: 2.384375 | lrm: 0.08 | dt: 1196.57ms | tok/sec: 876,319 | bf16_mfu: 53.83 | total time: 108.73m | eta: 2.0m step 05467/05568 (98.19%) | loss: 2.386639 | lrm: 0.08 | dt: 1188.57ms | tok/sec: 882,214 | bf16_mfu: 54.19 | total time: 108.75m | eta: 2.0m step 05468/05568 (98.20%) | loss: 2.382052 | lrm: 0.08 | dt: 1193.46ms | tok/sec: 878,599 | bf16_mfu: 53.97 | total time: 108.77m | eta: 2.0m step 05469/05568 (98.22%) | loss: 2.375801 | lrm: 0.08 | dt: 1198.02ms | tok/sec: 875,256 | bf16_mfu: 53.76 | total time: 108.79m | eta: 2.0m step 05470/05568 (98.24%) | loss: 2.383235 | lrm: 0.08 | dt: 1197.35ms | tok/sec: 875,746 | bf16_mfu: 53.80 | total time: 108.81m | eta: 1.9m step 05471/05568 (98.26%) | loss: 2.385878 | lrm: 0.08 | dt: 1195.59ms | tok/sec: 877,035 | bf16_mfu: 53.87 | total time: 108.83m | eta: 1.9m step 05472/05568 (98.28%) | loss: 2.377836 | lrm: 0.08 | dt: 1195.99ms | tok/sec: 876,740 | bf16_mfu: 53.86 | total time: 108.85m | eta: 1.9m step 05473/05568 (98.29%) | loss: 2.376396 | lrm: 0.07 | dt: 1198.00ms | tok/sec: 875,269 | bf16_mfu: 53.77 | total time: 108.87m | eta: 1.9m step 05474/05568 (98.31%) | loss: 2.380695 | lrm: 0.07 | dt: 1199.32ms | tok/sec: 874,308 | bf16_mfu: 53.71 | total time: 108.89m | eta: 1.9m step 05475/05568 (98.33%) | loss: 2.386045 | lrm: 0.07 | dt: 1197.28ms | tok/sec: 875,797 | bf16_mfu: 53.80 | total time: 108.91m | eta: 1.8m step 05476/05568 (98.35%) | loss: 2.398179 | lrm: 0.07 | dt: 1190.48ms | tok/sec: 880,801 | bf16_mfu: 54.11 | total time: 108.93m | eta: 1.8m step 05477/05568 (98.37%) | loss: 2.404721 | lrm: 0.07 | dt: 1199.34ms | tok/sec: 874,297 | bf16_mfu: 53.71 | total time: 108.95m | eta: 1.8m step 05478/05568 (98.38%) | loss: 2.380633 | lrm: 0.07 | dt: 1194.46ms | tok/sec: 877,864 | bf16_mfu: 53.93 | total time: 108.97m | eta: 1.8m step 05479/05568 (98.40%) | loss: 2.376591 | lrm: 0.07 | dt: 1195.24ms | tok/sec: 877,291 | bf16_mfu: 53.89 | total time: 108.99m | eta: 1.8m step 05480/05568 (98.42%) | loss: 2.385869 | lrm: 0.07 | dt: 1191.52ms | tok/sec: 880,033 | bf16_mfu: 54.06 | total time: 109.01m | eta: 1.7m step 05481/05568 (98.44%) | loss: 2.389295 | lrm: 0.07 | dt: 1195.51ms | tok/sec: 877,094 | bf16_mfu: 53.88 | total time: 109.03m | eta: 1.7m step 05482/05568 (98.46%) | loss: 2.379405 | lrm: 0.07 | dt: 1193.40ms | tok/sec: 878,643 | bf16_mfu: 53.97 | total time: 109.05m | eta: 1.7m step 05483/05568 (98.47%) | loss: 2.370828 | lrm: 0.07 | dt: 1192.42ms | tok/sec: 879,367 | bf16_mfu: 54.02 | total time: 109.07m | eta: 1.7m step 05484/05568 (98.49%) | loss: 2.370317 | lrm: 0.07 | dt: 1197.03ms | tok/sec: 875,983 | bf16_mfu: 53.81 | total time: 109.09m | eta: 1.7m step 05485/05568 (98.51%) | loss: 2.383650 | lrm: 0.07 | dt: 1193.91ms | tok/sec: 878,270 | bf16_mfu: 53.95 | total time: 109.11m | eta: 1.6m step 05486/05568 (98.53%) | loss: 2.381176 | lrm: 0.07 | dt: 1196.01ms | tok/sec: 876,730 | bf16_mfu: 53.86 | total time: 109.12m | eta: 1.6m step 05487/05568 (98.55%) | loss: 2.381903 | lrm: 0.07 | dt: 1194.79ms | tok/sec: 877,625 | bf16_mfu: 53.91 | total time: 109.14m | eta: 1.6m step 05488/05568 (98.56%) | loss: 2.377283 | lrm: 0.07 | dt: 1190.80ms | tok/sec: 880,567 | bf16_mfu: 54.09 | total time: 109.16m | eta: 1.6m step 05489/05568 (98.58%) | loss: 2.380572 | lrm: 0.07 | dt: 1194.53ms | tok/sec: 877,811 | bf16_mfu: 53.92 | total time: 109.18m | eta: 1.6m step 05490/05568 (98.60%) | loss: 2.381209 | lrm: 0.07 | dt: 1193.44ms | tok/sec: 878,613 | bf16_mfu: 53.97 | total time: 109.20m | eta: 1.5m step 05491/05568 (98.62%) | loss: 2.400030 | lrm: 0.07 | dt: 1196.84ms | tok/sec: 876,118 | bf16_mfu: 53.82 | total time: 109.22m | eta: 1.5m step 05492/05568 (98.64%) | loss: 2.397097 | lrm: 0.07 | dt: 1194.36ms | tok/sec: 877,938 | bf16_mfu: 53.93 | total time: 109.24m | eta: 1.5m step 05493/05568 (98.65%) | loss: 2.397914 | lrm: 0.07 | dt: 1196.20ms | tok/sec: 876,592 | bf16_mfu: 53.85 | total time: 109.26m | eta: 1.5m step 05494/05568 (98.67%) | loss: 2.391821 | lrm: 0.07 | dt: 1197.02ms | tok/sec: 875,989 | bf16_mfu: 53.81 | total time: 109.28m | eta: 1.5m step 05495/05568 (98.69%) | loss: 2.385276 | lrm: 0.07 | dt: 1196.03ms | tok/sec: 876,714 | bf16_mfu: 53.85 | total time: 109.30m | eta: 1.4m step 05496/05568 (98.71%) | loss: 2.375008 | lrm: 0.07 | dt: 1194.58ms | tok/sec: 877,776 | bf16_mfu: 53.92 | total time: 109.32m | eta: 1.4m step 05497/05568 (98.72%) | loss: 2.374722 | lrm: 0.07 | dt: 1194.25ms | tok/sec: 878,022 | bf16_mfu: 53.93 | total time: 109.34m | eta: 1.4m step 05498/05568 (98.74%) | loss: 2.383857 | lrm: 0.07 | dt: 1196.23ms | tok/sec: 876,569 | bf16_mfu: 53.85 | total time: 109.36m | eta: 1.4m step 05499/05568 (98.76%) | loss: 2.365140 | lrm: 0.07 | dt: 1196.64ms | tok/sec: 876,268 | bf16_mfu: 53.83 | total time: 109.38m | eta: 1.4m step:5500/5568 val_bpb:0.719696 val_time:3.90s step 05500/05568 (98.78%) | loss: 2.362243 | lrm: 0.07 | dt: 1192.33ms | tok/sec: 879,437 | bf16_mfu: 54.02 | total time: 109.40m | eta: 1.3m step 05501/05568 (98.80%) | loss: 2.361293 | lrm: 0.07 | dt: 1194.14ms | tok/sec: 878,098 | bf16_mfu: 53.94 | total time: 109.42m | eta: 1.3m step 05502/05568 (98.81%) | loss: 2.364188 | lrm: 0.07 | dt: 1197.88ms | tok/sec: 875,356 | bf16_mfu: 53.77 | total time: 109.44m | eta: 1.3m step 05503/05568 (98.83%) | loss: 2.366699 | lrm: 0.07 | dt: 1195.92ms | tok/sec: 876,796 | bf16_mfu: 53.86 | total time: 109.46m | eta: 1.3m step 05504/05568 (98.85%) | loss: 2.364614 | lrm: 0.07 | dt: 1191.28ms | tok/sec: 880,211 | bf16_mfu: 54.07 | total time: 109.48m | eta: 1.3m step 05505/05568 (98.87%) | loss: 2.370716 | lrm: 0.07 | dt: 1192.69ms | tok/sec: 879,165 | bf16_mfu: 54.01 | total time: 109.50m | eta: 1.2m step 05506/05568 (98.89%) | loss: 2.373079 | lrm: 0.07 | dt: 1198.94ms | tok/sec: 874,584 | bf16_mfu: 53.72 | total time: 109.52m | eta: 1.2m step 05507/05568 (98.90%) | loss: 2.375267 | lrm: 0.07 | dt: 1193.27ms | tok/sec: 878,742 | bf16_mfu: 53.98 | total time: 109.54m | eta: 1.2m step 05508/05568 (98.92%) | loss: 2.375991 | lrm: 0.07 | dt: 1196.55ms | tok/sec: 876,334 | bf16_mfu: 53.83 | total time: 109.56m | eta: 1.2m step 05509/05568 (98.94%) | loss: 2.369011 | lrm: 0.07 | dt: 1195.30ms | tok/sec: 877,252 | bf16_mfu: 53.89 | total time: 109.58m | eta: 1.2m step 05510/05568 (98.96%) | loss: 2.371911 | lrm: 0.07 | dt: 1191.87ms | tok/sec: 879,772 | bf16_mfu: 54.04 | total time: 109.60m | eta: 1.1m step 05511/05568 (98.98%) | loss: 2.370721 | lrm: 0.06 | dt: 1196.69ms | tok/sec: 876,231 | bf16_mfu: 53.82 | total time: 109.62m | eta: 1.1m step 05512/05568 (98.99%) | loss: 2.357878 | lrm: 0.06 | dt: 1199.09ms | tok/sec: 874,478 | bf16_mfu: 53.72 | total time: 109.64m | eta: 1.1m step 05513/05568 (99.01%) | loss: 2.360773 | lrm: 0.06 | dt: 1190.95ms | tok/sec: 880,451 | bf16_mfu: 54.08 | total time: 109.66m | eta: 1.1m step 05514/05568 (99.03%) | loss: 2.364379 | lrm: 0.06 | dt: 1195.84ms | tok/sec: 876,850 | bf16_mfu: 53.86 | total time: 109.68m | eta: 1.1m step 05515/05568 (99.05%) | loss: 2.371432 | lrm: 0.06 | dt: 1195.60ms | tok/sec: 877,027 | bf16_mfu: 53.87 | total time: 109.70m | eta: 1.0m step 05516/05568 (99.07%) | loss: 2.372191 | lrm: 0.06 | dt: 1194.85ms | tok/sec: 877,582 | bf16_mfu: 53.91 | total time: 109.72m | eta: 1.0m step 05517/05568 (99.08%) | loss: 2.386117 | lrm: 0.06 | dt: 1196.63ms | tok/sec: 876,271 | bf16_mfu: 53.83 | total time: 109.74m | eta: 1.0m step 05518/05568 (99.10%) | loss: 2.385363 | lrm: 0.06 | dt: 1195.06ms | tok/sec: 877,423 | bf16_mfu: 53.90 | total time: 109.76m | eta: 1.0m step 05519/05568 (99.12%) | loss: 2.387823 | lrm: 0.06 | dt: 1195.31ms | tok/sec: 877,243 | bf16_mfu: 53.89 | total time: 109.78m | eta: 1.0m step 05520/05568 (99.14%) | loss: 2.390370 | lrm: 0.06 | dt: 1193.16ms | tok/sec: 878,825 | bf16_mfu: 53.98 | total time: 109.80m | eta: 0.9m step 05521/05568 (99.16%) | loss: 2.392526 | lrm: 0.06 | dt: 1197.36ms | tok/sec: 875,738 | bf16_mfu: 53.79 | total time: 109.82m | eta: 0.9m step 05522/05568 (99.17%) | loss: 2.391632 | lrm: 0.06 | dt: 1193.64ms | tok/sec: 878,470 | bf16_mfu: 53.96 | total time: 109.84m | eta: 0.9m step 05523/05568 (99.19%) | loss: 2.384968 | lrm: 0.06 | dt: 1192.13ms | tok/sec: 879,585 | bf16_mfu: 54.03 | total time: 109.86m | eta: 0.9m step 05524/05568 (99.21%) | loss: 2.385077 | lrm: 0.06 | dt: 1194.15ms | tok/sec: 878,093 | bf16_mfu: 53.94 | total time: 109.88m | eta: 0.9m step 05525/05568 (99.23%) | loss: 2.394269 | lrm: 0.06 | dt: 1198.27ms | tok/sec: 875,073 | bf16_mfu: 53.75 | total time: 109.90m | eta: 0.8m step 05526/05568 (99.25%) | loss: 2.389792 | lrm: 0.06 | dt: 1196.08ms | tok/sec: 876,676 | bf16_mfu: 53.85 | total time: 109.92m | eta: 0.8m step 05527/05568 (99.26%) | loss: 2.377413 | lrm: 0.06 | dt: 1196.07ms | tok/sec: 876,686 | bf16_mfu: 53.85 | total time: 109.94m | eta: 0.8m step 05528/05568 (99.28%) | loss: 2.376661 | lrm: 0.06 | dt: 1195.45ms | tok/sec: 877,135 | bf16_mfu: 53.88 | total time: 109.96m | eta: 0.8m step 05529/05568 (99.30%) | loss: 2.373891 | lrm: 0.06 | dt: 1191.48ms | tok/sec: 880,058 | bf16_mfu: 54.06 | total time: 109.98m | eta: 0.8m step 05530/05568 (99.32%) | loss: 2.373368 | lrm: 0.06 | dt: 1203.90ms | tok/sec: 870,986 | bf16_mfu: 53.50 | total time: 110.00m | eta: 0.7m step 05531/05568 (99.34%) | loss: 2.365501 | lrm: 0.06 | dt: 1196.54ms | tok/sec: 876,340 | bf16_mfu: 53.83 | total time: 110.02m | eta: 0.7m step 05532/05568 (99.35%) | loss: 2.357026 | lrm: 0.06 | dt: 1190.77ms | tok/sec: 880,589 | bf16_mfu: 54.09 | total time: 110.04m | eta: 0.7m step 05533/05568 (99.37%) | loss: 2.352914 | lrm: 0.06 | dt: 1197.45ms | tok/sec: 875,675 | bf16_mfu: 53.79 | total time: 110.06m | eta: 0.7m step 05534/05568 (99.39%) | loss: 2.357845 | lrm: 0.06 | dt: 1199.52ms | tok/sec: 874,161 | bf16_mfu: 53.70 | total time: 110.08m | eta: 0.7m step 05535/05568 (99.41%) | loss: 2.358677 | lrm: 0.06 | dt: 1191.53ms | tok/sec: 880,024 | bf16_mfu: 54.06 | total time: 110.10m | eta: 0.6m step 05536/05568 (99.43%) | loss: 2.367057 | lrm: 0.06 | dt: 1191.96ms | tok/sec: 879,708 | bf16_mfu: 54.04 | total time: 110.12m | eta: 0.6m step 05537/05568 (99.44%) | loss: 2.366851 | lrm: 0.06 | dt: 1196.42ms | tok/sec: 876,430 | bf16_mfu: 53.84 | total time: 110.14m | eta: 0.6m step 05538/05568 (99.46%) | loss: 2.374662 | lrm: 0.06 | dt: 1195.84ms | tok/sec: 876,849 | bf16_mfu: 53.86 | total time: 110.16m | eta: 0.6m step 05539/05568 (99.48%) | loss: 2.371672 | lrm: 0.06 | dt: 1192.37ms | tok/sec: 879,406 | bf16_mfu: 54.02 | total time: 110.18m | eta: 0.6m step 05540/05568 (99.50%) | loss: 2.383900 | lrm: 0.06 | dt: 1195.73ms | tok/sec: 876,935 | bf16_mfu: 53.87 | total time: 110.20m | eta: 0.5m step 05541/05568 (99.52%) | loss: 2.380591 | lrm: 0.06 | dt: 1193.10ms | tok/sec: 878,869 | bf16_mfu: 53.99 | total time: 110.22m | eta: 0.5m step 05542/05568 (99.53%) | loss: 2.375645 | lrm: 0.06 | dt: 1191.24ms | tok/sec: 880,237 | bf16_mfu: 54.07 | total time: 110.24m | eta: 0.5m step 05543/05568 (99.55%) | loss: 2.371925 | lrm: 0.06 | dt: 1197.67ms | tok/sec: 875,516 | bf16_mfu: 53.78 | total time: 110.26m | eta: 0.5m step 05544/05568 (99.57%) | loss: 2.362219 | lrm: 0.06 | dt: 1191.74ms | tok/sec: 879,871 | bf16_mfu: 54.05 | total time: 110.28m | eta: 0.5m step 05545/05568 (99.59%) | loss: 2.372142 | lrm: 0.06 | dt: 1192.64ms | tok/sec: 879,202 | bf16_mfu: 54.01 | total time: 110.30m | eta: 0.4m step 05546/05568 (99.60%) | loss: 2.368170 | lrm: 0.06 | dt: 1194.73ms | tok/sec: 877,668 | bf16_mfu: 53.91 | total time: 110.32m | eta: 0.4m step 05547/05568 (99.62%) | loss: 2.376481 | lrm: 0.06 | dt: 1192.66ms | tok/sec: 879,190 | bf16_mfu: 54.01 | total time: 110.34m | eta: 0.4m step 05548/05568 (99.64%) | loss: 2.373384 | lrm: 0.06 | dt: 1195.06ms | tok/sec: 877,423 | bf16_mfu: 53.90 | total time: 110.36m | eta: 0.4m step 05549/05568 (99.66%) | loss: 2.378889 | lrm: 0.05 | dt: 1196.46ms | tok/sec: 876,400 | bf16_mfu: 53.84 | total time: 110.38m | eta: 0.4m step 05550/05568 (99.68%) | loss: 2.375448 | lrm: 0.05 | dt: 1192.78ms | tok/sec: 879,100 | bf16_mfu: 54.00 | total time: 110.40m | eta: 0.3m step 05551/05568 (99.69%) | loss: 2.378893 | lrm: 0.05 | dt: 1195.00ms | tok/sec: 877,468 | bf16_mfu: 53.90 | total time: 110.42m | eta: 0.3m step 05552/05568 (99.71%) | loss: 2.384237 | lrm: 0.05 | dt: 1197.00ms | tok/sec: 876,002 | bf16_mfu: 53.81 | total time: 110.44m | eta: 0.3m step 05553/05568 (99.73%) | loss: 2.383319 | lrm: 0.05 | dt: 1193.42ms | tok/sec: 878,627 | bf16_mfu: 53.97 | total time: 110.46m | eta: 0.3m step 05554/05568 (99.75%) | loss: 2.383341 | lrm: 0.05 | dt: 1195.34ms | tok/sec: 877,216 | bf16_mfu: 53.89 | total time: 110.48m | eta: 0.3m step 05555/05568 (99.77%) | loss: 2.386868 | lrm: 0.05 | dt: 1195.86ms | tok/sec: 876,838 | bf16_mfu: 53.86 | total time: 110.50m | eta: 0.2m step 05556/05568 (99.78%) | loss: 2.383745 | lrm: 0.05 | dt: 1200.68ms | tok/sec: 873,317 | bf16_mfu: 53.65 | total time: 110.52m | eta: 0.2m step 05557/05568 (99.80%) | loss: 2.384292 | lrm: 0.05 | dt: 1198.02ms | tok/sec: 875,261 | bf16_mfu: 53.77 | total time: 110.54m | eta: 0.2m step 05558/05568 (99.82%) | loss: 2.378706 | lrm: 0.05 | dt: 1194.64ms | tok/sec: 877,736 | bf16_mfu: 53.92 | total time: 110.56m | eta: 0.2m step 05559/05568 (99.84%) | loss: 2.389061 | lrm: 0.05 | dt: 1194.70ms | tok/sec: 877,689 | bf16_mfu: 53.91 | total time: 110.58m | eta: 0.2m step 05560/05568 (99.86%) | loss: 2.383528 | lrm: 0.05 | dt: 1193.48ms | tok/sec: 878,587 | bf16_mfu: 53.97 | total time: 110.60m | eta: 0.1m step 05561/05568 (99.87%) | loss: 2.389306 | lrm: 0.05 | dt: 1194.35ms | tok/sec: 877,945 | bf16_mfu: 53.93 | total time: 110.62m | eta: 0.1m step 05562/05568 (99.89%) | loss: 2.382024 | lrm: 0.05 | dt: 1195.96ms | tok/sec: 876,764 | bf16_mfu: 53.86 | total time: 110.64m | eta: 0.1m step 05563/05568 (99.91%) | loss: 2.382971 | lrm: 0.05 | dt: 1191.15ms | tok/sec: 880,307 | bf16_mfu: 54.08 | total time: 110.66m | eta: 0.1m step 05564/05568 (99.93%) | loss: 2.380679 | lrm: 0.05 | dt: 1196.22ms | tok/sec: 876,572 | bf16_mfu: 53.85 | total time: 110.68m | eta: 0.1m step 05565/05568 (99.95%) | loss: 2.383385 | lrm: 0.05 | dt: 1197.42ms | tok/sec: 875,698 | bf16_mfu: 53.79 | total time: 110.70m | eta: 0.0m step 05566/05568 (99.96%) | loss: 2.387411 | lrm: 0.05 | dt: 1191.85ms | tok/sec: 879,791 | bf16_mfu: 54.04 | total time: 110.72m | eta: 0.0m step 05567/05568 (99.98%) | loss: 2.388180 | lrm: 0.05 | dt: 1198.08ms | tok/sec: 875,210 | bf16_mfu: 53.76 | total time: 110.74m | eta: 0.0m step:5568/5568 val_bpb:0.719042 val_time:3.89s checkpoint captured at step 5568 (16.9s) Evaluating: hellaswag_zeroshot (multiple_choice, 10042 examples)... accuracy: 0.5523 | centered: 0.4030 | time: 3.52s Evaluating: jeopardy (language_modeling, 2117 examples)... accuracy: 0.0888 | centered: 0.0888 | time: 0.55s Evaluating: bigbench_qa_wikidata (language_modeling, 20321 examples)... accuracy: 0.4751 | centered: 0.4751 | time: 1.59s Evaluating: arc_easy (multiple_choice, 2376 examples)... accuracy: 0.6936 | centered: 0.5915 | time: 1.72s Evaluating: arc_challenge (multiple_choice, 1172 examples)... accuracy: 0.3925 | centered: 0.1900 | time: 0.96s Evaluating: copa (multiple_choice, 100 examples)... accuracy: 0.6200 | centered: 0.2400 | time: 0.34s Evaluating: commonsense_qa (multiple_choice, 1221 examples)... accuracy: 0.2424 | centered: 0.0530 | time: 1.19s Evaluating: piqa (multiple_choice, 1838 examples)... accuracy: 0.7524 | centered: 0.5049 | time: 0.72s Evaluating: openbook_qa (multiple_choice, 500 examples)... accuracy: 0.4060 | centered: 0.2080 | time: 0.33s Evaluating: lambada_openai (language_modeling, 5153 examples)... accuracy: 0.4372 | centered: 0.4372 | time: 0.25s Evaluating: hellaswag (multiple_choice, 10042 examples)... accuracy: 0.5555 | centered: 0.4073 | time: 16.38s Evaluating: winograd (schema, 273 examples)... accuracy: 0.6703 | centered: 0.3407 | time: 0.67s Evaluating: winogrande (schema, 1267 examples)... accuracy: 0.5635 | centered: 0.1271 | time: 0.30s Evaluating: bigbench_dyck_languages (language_modeling, 1000 examples)... accuracy: 0.1380 | centered: 0.1380 | time: 0.52s Evaluating: agi_eval_lsat_ar (multiple_choice, 230 examples)... accuracy: 0.2522 | centered: 0.0652 | time: 0.47s Evaluating: bigbench_cs_algorithms (language_modeling, 1320 examples)... accuracy: 0.4379 | centered: 0.4379 | time: 0.52s Evaluating: bigbench_operators (language_modeling, 210 examples)... accuracy: 0.1857 | centered: 0.1857 | time: 0.29s Evaluating: bigbench_repeat_copy_logic (language_modeling, 32 examples)... accuracy: 0.0000 | centered: 0.0000 | time: 0.32s Evaluating: squad (language_modeling, 10570 examples)... accuracy: 0.1940 | centered: 0.1940 | time: 10.16s Evaluating: coqa (language_modeling, 7983 examples)... accuracy: 0.3088 | centered: 0.3088 | time: 2.53s Evaluating: boolq (multiple_choice, 3270 examples)... accuracy: 0.6076 | centered: -0.0325 | time: 5.13s Evaluating: bigbench_language_identification (multiple_choice, 10000 examples)... accuracy: 0.2499 | centered: 0.1748 | time: 31.38s CORE metric: 0.2517 | total CORE eval time: 80.46s hellaswag_zeroshot: accuracy=0.5523 centered=0.4030 jeopardy: accuracy=0.0888 centered=0.0888 bigbench_qa_wikidata: accuracy=0.4751 centered=0.4751 arc_easy: accuracy=0.6936 centered=0.5915 arc_challenge: accuracy=0.3925 centered=0.1900 copa: accuracy=0.6200 centered=0.2400 commonsense_qa: accuracy=0.2424 centered=0.0530 piqa: accuracy=0.7524 centered=0.5049 openbook_qa: accuracy=0.4060 centered=0.2080 lambada_openai: accuracy=0.4372 centered=0.4372 hellaswag: accuracy=0.5555 centered=0.4073 winograd: accuracy=0.6703 centered=0.3407 winogrande: accuracy=0.5635 centered=0.1271 bigbench_dyck_languages: accuracy=0.1380 centered=0.1380 agi_eval_lsat_ar: accuracy=0.2522 centered=0.0652 bigbench_cs_algorithms: accuracy=0.4379 centered=0.4379 bigbench_operators: accuracy=0.1857 centered=0.1857 bigbench_repeat_copy_logic: accuracy=0.0000 centered=0.0000 squad: accuracy=0.1940 centered=0.1940 coqa: accuracy=0.3088 centered=0.3088 boolq: accuracy=0.6076 centered=-0.0325 bigbench_language_identification: accuracy=0.2499 centered=0.1748 peak memory allocated: 71569 MiB reserved: 72736 MiB total training time: 110.74m minimum validation bpb: 0.719042