Wing Lian a27b909c5c GRPO fixes (peft) (#2676) ...

* don't set peft_config on grpo to prevent double peft wrap

* remove overrides needed to support bug

* fix grpo tests

* require more CPU for multigpu to help with torch compile for vllm

2025-05-16 15:47:03 -04:00

..

patched

SP GRPO support + batch SP fixes (#2643)

2025-05-12 17:52:40 -04:00

solo

GRPO fixes (peft) (#2676)

2025-05-16 15:47:03 -04:00

__init__.py

Attempt to run multigpu in PR CI for now to ensure it works (#1815) [skip ci]

2024-08-09 11:50:13 -04:00

test_eval.py

Updates for trl 0.16.0 - mostly for GRPO (#2437) [skip ci]

2025-03-31 15:47:11 -04:00

test_gemma3.py

gemma3 packing fixes (#2449)

2025-03-31 17:15:23 -04:00

test_llama.py

swap tinymodels that have safetensors for some ci tests (#2641)

2025-05-07 15:06:07 -04:00

test_qwen2.py

Update dependencies and show slow tests in CI (#2492)

2025-04-05 17:41:31 -04:00

test_ray.py

make e2e tests a bit faster by reducing test split size (#2522) [skip ci]

2025-04-12 07:24:43 -07:00