The new version (as of now) still has a comment making it obvious that there's an intentionally introduced bug, but it's not as on the nose: https://github.com/openai/SWELancer-Benchmark/blob/2a77e3572...
The new version (as of now) still has a comment making it obvious that there's an intentionally introduced bug, but it's not as on the nose: https://github.com/openai/SWELancer-Benchmark/blob/2a77e3572...