Open
Conversation
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
The actions for evals have been failing sine 2/13
In order to get some reliable evals for this project, I wanted to fix it.
Solution
I believe the core issue was feeding too much data into the claude
Readtool at once. ARC can produce 2-3MB, which seems to exceed the limit for claude, and their harness propagates that to the main loop.Proof that it now runs is here: https://github.com/maxtheman/node-rlm/actions/runs/23169689550
I didn't want to start catching issues randomly, so instead added logic to break the offending json file into several, and author a manifest.
misc
Happy to squash these commits or make any requested style or substance changes