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

Closed hijkzzz closed 4 months ago

hijkzzz commented 4 months ago

217 use right pad to reduce precision issues

hijkzzz commented 4 months ago

for this MR, there are some bugs: