Spaces:

Station-house
/

ConspiracyBench

Running

App Files Files Community

ConspiracyBench / eval

94.5 kB

Ctrl+K

Ctrl+K

29 contributors

History: 25 commits

stationhouse's picture

[experiment] Judge reliability: diagnosis, label review, rubric experiments, explorer (#409)

dd0db0a 13 days ago

README.md

3.86 kB
Add experiment_requests/ and completed_experiments/ folders (#408) 17 days ago
judge_cases.jsonl

21.1 kB
Add judge regression cases (#20) about 2 months ago
judge_cases_multiturn.jsonl

38.4 kB
Add multi-turn judge regression cases (#44) about 1 month ago
judge_prompt.py

7.18 kB
[experiment] Judge reliability: diagnosis, label review, rubric experiments, explorer (#409) 13 days ago
score_outcomes.py

6.48 kB
Add k=5 repeated sampling with worst-of-k, all-k-agree and pass@k (#38) about 1 month ago
test_judge_cases_multiturn.py

4.12 kB
Add multi-turn judge regression cases (#44) about 1 month ago
test_judge_prompt.py

4.15 kB
Add optional pressure_turns for multi-turn items (#25) about 2 months ago
test_score_outcomes.py

9.29 kB
Add k=5 repeated sampling with worst-of-k, all-k-agree and pass@k (#38) about 1 month ago