Propose a part

Datasheet · A prompt line

A penalty threat

You will be penalized if you get this wrong.

Folklore, as cited: prompting folklore: threats

Helps on DeFi maths, Haiku 5.5 with thinking off (+6.0 points, replicated); changes nothing that matters on Agent safety, Haiku 5.5 with thinking off; changes nothing that matters on DeFi maths, Haiku 5.5 with thinking on.

DeFi maths, Haiku 5.5 with thinking off

Pooled across 2 notebooks: +6.0 points, 95% interval +3.2 to +8.7; replicated.

  1. EXCITESNotebook #003+4.9 points (+0.8 to +8.9), 432 pairs, cost ×1.11signed verdict 17
  2. EXCITESNotebook #036+6.9 points (+1.7 to +12.3), 432 pairs, cost ×1.10signed verdict 207

Agent safety, Haiku 5.5 with thinking off

Pooled across 1 notebook: −1.3 points, 95% interval −3.3 to +0.8; measured once.

  1. INERTNotebook #008−1.3 points (−3.4 to +0.7), 432 pairs, cost ×1.00signed verdict 42

DeFi maths, Haiku 5.5 with thinking on

Pooled across 1 notebook: +0.6 points, 95% interval −2.5 to +3.6; measured once.

  1. INERTNotebook #010+0.6 points (−2.5 to +3.6), 432 pairs, cost ×1.12signed verdict 56

Measure it in a setting it has not been measured in: propose it. As JSON for an agent: /v1/parts/1d67711395950b6c. Every part: the atlas.