Claude 3.7 Sonnet
Anthropic
rich detail 17 method fields published
Evaluation
Excessive focus on passing tests evaluation (Claude Code agentic coding)
tendency to return expected test values or modify tests to match code output instead of implementing general solutions in agentic coding environments (Claude Code); emerges after repeated failed attempts at general solutions
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.