HumanSignal / RLHF

Collection of links, tutorials and best practices of how to collect the data and build end-to-end RLHF system to finetune Generative AI models
154 stars 31 forks source link

REWARD_CHECKPOINT_PATH ( How I solve following issue?) #7

Open nadimkaysar opened 1 month ago

nadimkaysar commented 1 month ago

Where did I find this 'REWARD_CHECKPOINT_PATH' as a bin file?
`
3 rw_tokenizer.pad_token = rw_tokenizer.eos_token

4 rw_model = GPTRewardModel(SFT_MODEL_PATH)

----> 5 rw_model.load_state_dict(REWARD_CHECKPOINT_PATH)

6 rw_model.half()

7 rw_model.eval() `