Gahow Wang 22b7434b23 test: AdamW PyTorch parity + checkpoint round-trip + real training ...

Acceptance tests (GPU-gated not(no_cuda), run on dash5):
- adamw_parity_dump.rs + adamw_parity.py: build the tiny model with fixed init,
  run N AdamW steps on a fixed batch, dump the loss trajectory + final params;
  the Python side rebuilds the identical model and runs torch.optim.AdamW with
  matched lr/wd/betas/eps, comparing trajectory + final params within rtol.
- checkpoint_roundtrip.rs: train a few steps, save, load into a fresh model with
  a DIFFERENT init, assert identical logits/loss on a fixed input.
- real_training.rs (#[ignore], --release): train on TinyStories for a bounded
  budget; assert loss drops substantially and print greedy samples.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-15 16:30:06 +08:00

..

src

train: loop + checkpoint save/load + sampler + train binary

2026-06-15 16:29:58 +08:00

tests

test: AdamW PyTorch parity + checkpoint round-trip + real training

2026-06-15 16:30:06 +08:00

build.rs

data: gpt2 bpe via xserv-tokenizer + TinyStories corpus + lr schedule + grad clip

2026-06-15 16:29:32 +08:00

Cargo.toml

train: loop + checkpoint save/load + sampler + train binary

2026-06-15 16:29:58 +08:00