v0.0.2-alpha12
Reward Model evaluation (#2314)
Subclass dataset (follow-up for https://github.com/LAION-AI/Open-Assistant/pull/2671) (#2686)
add one .gitignore to resolve conflicts
Document message and token format (#2056)
remove CUDA_VISIBLE_DEVICES= which is user specific (#2295)
Flash attention support for Llama (#2277)