Thanks for your awesome work so that the community can train LLM on very long context! However, I find that in the preprocess function, line
and line:
|
target[cur_len : cur_len + instruction_len] = IGNORE_TOKEN_ID |
will set
target to:
[1, -100, -100, ...], with the first element is not ignored. I think Fastchat gives the correct code, which is first setting
target[:cur_len] = IGNORE_TOKEN_ID so the target will be
[-100, -100, -100, ...]. Am I right?
@DachengLi1
Thanks for your awesome work so that the community can train LLM on very long context! However, I find that in the
preprocessfunction, lineLongChat/longchat/train/fine_tune/train.py
Line 125 in a824bda
and line:
LongChat/longchat/train/fine_tune/train.py
Line 137 in a824bda
will set
targetto:[1, -100, -100, ...], with the first element is not ignored. I think Fastchat gives the correct code, which is first settingtarget[:cur_len] = IGNORE_TOKEN_IDso the target will be[-100, -100, -100, ...]. Am I right?@DachengLi1