gptbase_dom_hom_1000mb
GPT2 (base) trained on a set of languages where English dominates the training data English is dominant in the model training set (3000mb of English and 500mb (both with byte premium scaling) for all other languages) Balanced tokenizer training set (100mb with byte premium scaling for each language)
Training results
Based on 2 W&B runs named gptbase_dom_hom_1000mb in sjzahra-ri-se/huggingface.
Loss: 2 run(s); warmup: 2; GPU telemetry: 2 run(s).
Training hardware: Trained on a single H100 GPU.
Languages
Danish, English, Swedish, Norwegian, Icelandic
Training metadata
| Parameter | Value |
|---|---|
| Runs included | 2 |
| Training hardware | Trained on a single H100 GPU |
| Total epochs | 9.99997 |
| Total training steps | 357,849 |
| Total training FLOPs | 5.9842e+18 |
| Learning rate | 0.0001 |
| Train batch size | 64 |
| Eval batch size | 64 |
| Gradient accumulation | 1 |
| Weight decay | 0.01 |
| Warmup steps | 35784 |
| Warmup ratio | 0 |
| Optimizer | adamw_torch |
| Scheduler | linear |
| Max sequence length | 20 |
| Precision | float32 (1 runs); False (1 runs) |
| Seed | 43 |
| GPU utilization | mean 85.3%, peak 100.0% |
Generated 2026-11-17T17:25:03+00:00.
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support
Collection including NLP-RISE/gptbase_dom_hom_1000mb
Collection
Trained on on swe_latn, nob_latn, isl_latn, eng_latn, and dan_latn (homogeneous set 1) โข 6 items โข Updated
