-
Notifications
You must be signed in to change notification settings - Fork 11k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
server: fix reported top tokens for temperature 0 #7203
Merged
JohannesGaessler
merged 1 commit into
ggml-org:master
from
JohannesGaessler:sampling-fix-n_cons
May 11, 2024
Merged
server: fix reported top tokens for temperature 0 #7203
JohannesGaessler
merged 1 commit into
ggml-org:master
from
JohannesGaessler:sampling-fix-n_cons
May 11, 2024
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
ggerganov
approved these changes
May 11, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
bugfix
fixes an issue or bug
Review Complexity : Low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
server
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Outdated suggestions cannot be applied.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.
Fixes #7197 .
I had incorrectly assumed that with greedy sampling the
size
of thellama_token_data_array
would be 1. Instead it is the vocabulary size and I did not catch this during testing. There is also the issue that the tokens are not sorted at all which would cause issues when requesting only the probability of the top token with temperature 0. Edit: ... when keeping the current interpretation ofn_considered
.This PR renames
n_considered
ton_valid
and slightly changes the intended interpretation. It now represents the number of correct top tokens with correct probabilities. For temperature 0 this is set to 0 so that the server always fetches the top tokens.