If have HF model/tokenizer, use that instead of faketokenizer (tiktoken) since see too large differences and failures even with 250 token buffer, still of by another 350. #352
Conversation
This file contains hidden or 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
…en) since see too large differences and failures even with 250 token buffer, still of by another 350.
pseudotensor
force-pushed
the
controlhftokens
branch
from
2d3bc00 to
3ac421b
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
If have HF model/tokenizer, use that instead of faketokenizer (tiktoken) since see too large differences and failures even with 250 token buffer, still of by another 350.