OpenLLMAI / OpenRLHF

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & Mixtral)
https://openrlhf.readthedocs.io/
Apache License 2.0
1.73k stars 164 forks source link

fix gradient_checkpointing_kwargs bug #206

Closed wwxFromTju closed 5 months ago

wwxFromTju commented 5 months ago

see: https://github.com/huggingface/trl/pull/912 , https://github.com/huggingface/transformers/pull/26917 and https://github.com/huggingface/transformers/issues/26969