Remove the fixed eot_token mechanism for SFT
#927
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Not all pretrained LLMs use `<|endoftext|>` as the `eot_token`, therefore it's inappropriate to fix it.
Xingfu-Yi
requested review from
tjruwase,
awan-10,
eltonzheng,
duli2012 and
arashb
as code owners
tjruwase
approved these changes
Sep 20, 2024
|
Hi @arashb, @duli2012, @awan-10, @eltonzheng, I hope you're doing well. When you have a moment, could you kindly take a look at this PR? It has already received one approval, but it seems to be stuck and needs further reviews to move forward. Thank you so much in advance for your time and help. Best regards, |
Hi @Xingfu-Yi - we will work on getting this PR merged, sorry for the delay. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Background
Not all pretrained LLMs use
<|endoftext|>as theeot_token, therefore it's inappropriate to fix it.Changes
eot_token:args.end_of_conversation_token = "<|endoftext|>".eot_tokenwhich is<|endoftext|>by default. Users can manually set the token according to the pretrained model they use.