Add tool to evaluate NeuronModelForCausalLM perplexity #692
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
By default, transformers_neuronx model only return the logits for the last input token, i.e. those leading to the generated token. This modifies the default behaviour for models that support it (all but gpt2), but only when batch_size is 1. This does not have too much impact on performance and allows to evaluate the model perplexity.
This script can be used to evaluate the perplexity of a float or neuron LLM model using a configurable dataset (WikiText by default).
dacorvo
force-pushed
the
llm_neuron_perplexity
branch
from
7d0e876
to
3375842
Compare
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
Obsoleted by EleutherAI/lm-evaluation-harness#2314 |
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Outdated suggestions cannot be applied.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.
What does this PR do?
This adds a small script that can evaluate the perplexity of a standard pytorch or Neuron LLM on a configurable dataset (defaults to WikiText).
Results for some models: