Benchmark
AgenticInterpBench
What is AgenticInterpBench?
Evaluates language model agents on explaining components of transformer circuits, with 84 semi-synthetic circuits and 163 component-level annotations.
- Released
- 2026-06-23
- Evaluates
- General AI, Language & Knowledge, cs.AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
- Reported scores
- 0
Source provenance
- Original evidence https://arxiv.org/abs/2606.24026
AgenticInterpBench paper
No reported scores are on record for this benchmark yet.