feature: align hallucinated package named with outputs #1076
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
There was a problem hiding this comment.
What is this file used for?
There was a problem hiding this comment.
Used to add the Rust stdlib names to the entries in crates.io, in the Rust package hallucination detector
There was a problem hiding this comment.
I see it now, the location may be better organized as data/packagehallucination/rust/std_entires.txt or maybe should be added to the huggingface dataset with corresponding dates of addition based on when they became supported in rust. This will impact usage when #950 is ready to merge.
There was a problem hiding this comment.
Yeah, I was on the fence about a dir with a single file but consistency is good and yeah, that PR may bring in more things. Will move it.
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Outdated suggestions cannot be applied.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.
Previously,
packagehallucinationprobes would attach a list of the names of all hallucinated packages at detection time, but this was hard to link back to individual outputs. This PR makes the reporting of hallucinated packages a list of lists aligned with outputs.Example
given an output output:
["import not_a_real_package", "import sys", "pass"]after running
garak.detectors.packagehallucination.PythonPypi()on this,without PR:
attempt.notes["hallucinated_python_packages"]would be["not_a_real_package"]with PR:
attempt.notes["hallucinated_python_packages"]is[["not_a_real_package"], [None], []]Verification
List the steps needed to make sure this thing works