Oct 14 10 days ago 20:00 One arithmetic on every device: mixed CPU+GPU optimizer, replica hash check, FTZ pin, subnormal gate, live f32 audit, probe/fragment fixes (2026-10-14) Oct 1 23 days ago 05:39 Logic pass: CPU attention mirrors get the same int32 bound as the GPU kernels (peers refuse identically); gpuMlpChain returns pooled buffers when the row-max guard throws 05:01 Reductions refuse what they used to accept: GPU row-max non-finite guard, averageGrads length/NaN refusal, int32 accumulator bound, capacity-weighted cluster loss, finite reduced-gradient check, empty CUDA_VISIBLE_DEVICES Sep 27 27 days ago 12:26 web: restore LF line endings on verified_core.js / transformer.js The previous upload carried CRLF line endings on these two files (a local git checkout artifact). Content is otherwise unchanged; this restores the repo's LF convention so the files diff cleanly. 12:19 Refuse NaN/Inf at every int8 quantize; requant rounds half-up (retrained); bounded units; README for September Four local commits in one upload (ec432c8, c8db1cf, 47724cd, ae9cfa2): bounded elementwise units and LUT build (407 MB -> 2 MB, 277 MB -> 1 MB at startup); requant16 retrained from floor (-0.4981 LSB bias) to round-half-up, N/N 65,536/65,536, browser requant LUT re-exported; qat.py and every browser quantizer now refuse NaN/Inf instead of silently quantizing the whole layer to 0 (new tests fail on the old code); README September section. python test_verified_units.py 25/25, npm test 12/12 suites green. Jul 18 08:40 Upload web/public/app.js with huggingface_hub 08:40 Upload web/public/webgpu.js with huggingface_hub 08:29 Upload web/TEST_RESULTS.md with huggingface_hub 08:29 Upload web/test_gates.js with huggingface_hub 08:29 Upload web/public/webgpu.js with huggingface_hub 08:29 Upload web/public/transformer.js with huggingface_hub 08:29 Upload web/public/verified_core.js with huggingface_hub 07:59 Upload web/TEST_RESULTS.md with huggingface_hub 07:59 Upload web/test_gates.js with huggingface_hub 07:59 Upload web/public/app.js with huggingface_hub 07:59 Upload web/public/verified_core.js with huggingface_hub 07:47 Upload web/public/webgpu.js with huggingface_hub 07:42 Upload web/TEST_RESULTS.md with huggingface_hub 07:42 Upload web/public/webgpu.js with huggingface_hub 06:25 Upload web/TEST_RESULTS.md with huggingface_hub Jul 17 23:44 Upload web/TEST_RESULTS.md with huggingface_hub 23:44 Upload web/test_vectors_fp32.json with huggingface_hub 23:44 Upload web/test_ieee.js with huggingface_hub 22:19 Upload web/TEST_RESULTS.md with huggingface_hub 22:19 Upload web/public/webgpu.js with huggingface_hub 08:30 Upload web/TEST_RESULTS.md with huggingface_hub 08:30 Upload web/public/webgpu.js with huggingface_hub 08:16 Upload web/TEST_RESULTS.md with huggingface_hub 08:16 Upload web/test_vectors_fma.json with huggingface_hub 08:16 Upload web/test_ieee.js with huggingface_hub 07:36 Upload web/TEST_RESULTS.md with huggingface_hub 07:36 Upload web/package.json with huggingface_hub 07:36 Upload web/test_corpus.js with huggingface_hub 07:36 Upload web/test_selfcorpus.js with huggingface_hub 02:32 Upload web/README.md with huggingface_hub 02:32 Upload web/docs/TROUBLESHOOTING.md with huggingface_hub 02:32 Upload web/docs/VERIFICATION.md with huggingface_hub 02:32 Upload web/docs/ARCHITECTURE.md with huggingface_hub 02:32 Upload web/docs/GETTING_STARTED.md with huggingface_hub 02:06 Upload web/TEST_RESULTS.md with huggingface_hub 02:06 Upload web/test_b2b.js with huggingface_hub 02:06 Upload web/public/webgpu.js with huggingface_hub 01:52 Upload web/TEST_RESULTS.md with huggingface_hub 01:52 Upload web/test_gates.js with huggingface_hub 01:52 Upload web/public/verified_core.js with huggingface_hub 01:52 Upload web/public/webgpu.js with huggingface_hub Jul 16 20:52 web: property suite 4/4 on the external corpus (definitional absolutes join the relations); de-flaked test_b2b 20:38 Remove stray web/webgpu.js (belongs only at web/public/webgpu.js) 20:38 web: gate-the-gate boundary discrimination in the B2B init gate 10:50 web: logic-pass fixes (portable test_b2b, symmetric repair patience) 
