PKU-Alignment safe-rlhf Discussions
Pinned Discussions
Discussions
-
You must be logged in to vote -
You must be logged in to vote Better deepspeed configurations
enhancementNew feature or request help wantedExtra attention is needed questionFurther information is requested -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote -
You must be logged in to vote Loss/reward training curve.
questionFurther information is requested -
You must be logged in to vote Expected training time
cudaSomething CUDA related -
You must be logged in to vote Add
bugmin_new_tokens=1for generationSomething isn't working -
You must be logged in to vote tokenizer and reward_tokenizer have different special tokens
bugSomething isn't working