Fix MultinomialSampler hyperparameter bug #1150
Open
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
lapp0
reviewed
Sep 14, 2024
|
I initialized a sampler and passed it into a regex generator. sampler = outlines.samplers.MultinomialSampler(temperature=1.5, top_k=100, samples=5)
generator = outlines.generate.regex(
model,
r"I really love (\S*) cake with ice cream.",
sampler=sampler
)
structured = generator(prompt, max_tokens=30)First which calls which calls |
|
I assume this was with ExLlamaV2, or am I wrong? Good find. |
Uses all logit_processors instead of just the last logit_processor
rlouf
force-pushed
the
fix-mnomial-hparams
branch
from
50feaa4
to
c27a78e
Compare
|
I'm not sure about |
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Outdated suggestions cannot be applied.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.
Currently seems like the MultinomialSampler does not utilize all logit_processors, only the last one in self.logit_processors. Thus, some of the hyperparameters (i.e. top_k) are not utilized. This fixes this issue with a small variable change.