Spaces:

towardsai-tutors
/

ai-tutor-chatbot

Running

App Files Files Community

ai-tutor-chatbot / tests /test_evals_grade.py

Commit History

feat(evals): paired lockstep runner, lossless subagent grading, deprecated-battery guard

5b3529a

omarsol Claude Fable 5 commited on Jul 16

fix(evals): never judge faithfulness on silently truncated evidence

2ef4c7f

omarsol Claude Fable 5 commited on Jul 6

wip: eval harness + public docs bundle; relocate eval docs into evals/ (#10)

dc474cb
unverified

Omar Solano Claude Opus 4.8 (1M context) commited on Jun 29

Add evaluation harness for memory/context-management experiments

a04f9ea

omarsol Claude Fable 5 commited on Jun 12