Switch reference SUT from GPT2 to Pythia 70m #114
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
…of items used for calibration run.
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
mkly
reviewed
Feb 22, 2024
dhosterman
reviewed
Feb 23, 2024
| harms = GeneralChatBotBenchmarkDefinition().harms() | ||
| harm_scores = run_tests(harms, reference_sut, 45) | ||
| harm_scores = run_tests(harms, reference_sut, 100) |
There was a problem hiding this comment.
Is the increase in number of items here specific to pythia-70m or is there some other reason for this change?
There was a problem hiding this comment.
This increase happened because Pythia was causing fewer blowups, presumably because it generates fewer empty responses. Long term I'll shoot for something like 1000 examples per test in hopes that will give us more stable and reliable calibrations.
dhosterman
approved these changes
Feb 23, 2024
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
In service of #113.