Datasets:

S3RL4
/

binary-reasoning-v1

Modalities:

Formats:

Size:

Tags:

reinforcement-learning

reverse-engineering

Libraries:

License:

Dataset card Data Studio Files Files and versions

binary-reasoning-v1 / EVAL_RESULTS.md

Commit History

Nov 1 2 days ago

12:33
Public clean-up: measured public outcomes only; private families shared privately; licence clarity

S3RL4

Verified signature
e67a78d

Oct 30 4 days ago

06:31
Update EVAL_RESULTS.md: difficulty wording and caveats

S3RL4

Verified signature
45dd2b3

Oct 29 5 days ago

07:16
frontier reference scores (Opus 5.5 measured 2026-10-29)

S3RL4

Verified signature
62079a7

Oct 28 6 days ago

20:52
add baseline eval methodology + results

S3RL4

Verified signature
1c4be39