New tasks and evaluation code for LLMs #160
Merged
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
Add eval script with more verbose output and on-the-fly models
Logging updates
Add option to show a few model inputs
Fix missing file for num_model_inputs
Add full MC prompts for ARC
Adding summarization datasets, scitldr and xsum
Verbose perplexity evaluation
Fix prefix space issues with certain tokenizers
Fix bug in loading datasets
Add siqa and csqa tasks
Olmo eval cleanup
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Outdated suggestions cannot be applied.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 includes a number of additions associated with LLM evaluations:
New tasks under catwalk/dependencies/lm_eval: arc_easy:mc, arc_challenge:mc, case_hold:mc, csqa, eurlex, naturalqs_short_open, unfair_tos, scitldr, social_iqa, xsum
New detailed metrics for ranked classification in
RankedClassificationMetrics.New task for perplexity scoring over a set of jsonl files (
catwalk/tasks/perplexity_jsonl.py)New model type
"lm::"for the general types of tasks handled by current decoder-only LLMs (catwalk/models/language_model.py).Script for running LLM evaluations (
catwalk/run_lm_eval.py).