feat: add perplexity example #313
Draft
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
| # Create a tokenizer. | ||
| tokenizer = AutoTokenizer.from_pretrained(model_id) | ||
|
|
||
| # Tokenize the prompts and discard or truncate any prompts longer than 2048 tokens. |
There was a problem hiding this comment.
max_length is different below, maybe you meant 4096?
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Outdated suggestions cannot be applied.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.
This PR adds an example to calculate model perplexity, this approach performs it by using
prompt_logprobs:The last step should technically be unneeded, but to match the perplexity results from llama.cpp, we seem to need the division by two. It still seems unreliable, as Llama-2 7B shows a lower ppx than Mistral 7B, and FP8 KV Cache doesn't change ppx at all.