Open
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Previously, our logprobs multiple choice solvers worked by concatenating each choice with all previous input, and getting the log probability for the entire sequence for each choice. This has two disadvantages: we are processing the (sometimes lengthy) prefix multiple times, and the log probabilities that we get include the probability of the prefix.
This PR reworks the generation code so it happens in two stages: we first process the prefix by itself and save the resulting KV cache, then we separately process the choices (in a single batch) using that cache. This should be more performant and gives us logprobs for the choices independent of the prefix.