change to right-padding, update reward score #803

Former-commit-id: baa90415bc8f5ebd423d001378b51c3a3a6c2ec7
This commit is contained in:
hiyouga
2023-09-08 20:04:31 +08:00
parent bb1b67c076
commit 612d97db6f
15 changed files with 97 additions and 59 deletions

View File

@@ -1,5 +1,4 @@
# Inspired by:
# https://github.com/lvwerra/trl/blob/main/examples/summarization/scripts/reward_summarization.py
# https://github.com/CarperAI/trlx/blob/main/examples/summarize_rlhf/reward_model/train_reward_model_gptj.py
from typing import TYPE_CHECKING, Optional, List