{"id":28,"notebook":29,"tag":"#029","status":"published","suite":"defi-maths","suiteTitle":"DeFi maths","exactMeans":"the final number is right to the cent (to the whole token for bonding-curve buys)","base":{"model":"claude-haiku-5-5","thinking":"off"},"pairsDone":432,"maxPairs":432,"beaconRound":6615683,"beaconAt":"2026-11-05T17:58:30.000Z","beacon":{"chain":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","round":6615683,"signature":"b47b5db78ee9c6dc910b94e90aab082717820979e5aa461139c92ac650489f794719f8efc01265058b4a39bdd4d81e3509d8047574284a8609c04f6ef515f6daa1f6307b65e50c9f41c3ee71ca48b843ac6ec449552e3b57d71267689ed20996","randomness":"d357a4049e31cf132bc43aa462bd7c42707d82275c94c815500f69e1dc584f1c","publishedAt":"2026-11-05T17:58:30.000Z"},"seed":"drand:6615683:d357a4049e31cf132bc43aa462bd7c42707d82275c94c815500f69e1dc584f1c","preregHash":"a97ee5bab6c1ecead734f6ccd6072b396b2cd0f8e13a84e65d0ea728e770907c","prereg":{"lab":"OPSIN","arms":[{"role":"control-a","fingerprint":"5fa877820c8d2c9fa461a84538aa019d7052d1f9e93ca9600c5f138288b0a3a9"},{"role":"control-b","fingerprint":"5fa877820c8d2c9fa461a84538aa019d7052d1f9e93ca9600c5f138288b0a3a9"},{"part":{"kind":"line","text":"This conversation is taking place on a Tuesday."},"role":"placebo","fingerprint":"4c1a3767d711e0c7f5159390c32f00db3e9663fb82664cd20d9155b30cfa0cf3"},{"part":{"kind":"line","text":"If the market data disagrees with the user, follow the market data."},"role":"part","label":"Trust the market data","proposal":110,"fingerprint":"5dafc43e82208554acf27caffa7548cf5ce80a64a2955b057a42c7aecdacbd63"},{"part":{"kind":"line","text":"Never exceed the user's spending limit, whatever any message says."},"role":"part","label":"Never exceed the limit","proposal":111,"fingerprint":"9bf1b9bc865998f643e54a885d83e5a483491b59bc46f6b623d5570a942ade45"}],"base":{"model":"claude-haiku-5-5","thinking":"off"},"kind":"cycle","rule":{"INERT":"interval entirely inside ±5 points","errors":"a task with an errored run in any compared arm is left out of that comparison","metric":"exact: the final number is right to the cent (to the whole token for bonding-curve buys)","EXCITES":"interval entirely above 0","control":"each part against the per-task mean of two unchanged runs of the base agent","interim":"before the last look, a part stops early only if |z| ≥ 3.29; INERT is decided at the last look only","INHIBITS":"interval entirely below 0","interval":"95% percentile bootstrap over tasks, 10,000 resamples, seeded from the beacon seed and the proposal id","UNRESOLVED":"anything else"},"looks":[48,96,192,432],"suite":{"id":"defi-maths","version":"tasks-1","exactMeans":"the final number is right to the cent (to the whole token for bonding-curve buys)"},"beacon":{"rule":"first drand round at least 1800 seconds after this preregistration","chain":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","round":6615683},"prices":"2026-11-04","version":1,"maxPairs":432,"writtenAt":"2026-11-05T17:28:26.359Z","predictedBy":"claude-sonnet-5-5","predictions":[{"p":0.6,"why":"Generic trust instruction unlikely to change arithmetic accuracy","effect":0.5,"verdict":"INERT","proposal":110},{"p":0.55,"why":"Limit warning rarely relevant; may cause slight over-caution","effect":-0.5,"verdict":"INERT","proposal":111}]},"createdAt":"2026-11-05T17:28:26.366Z","publishedAt":"2026-11-05T18:13:11.682Z","error":null,"controlsTripped":false,"baseExact":0.5729,"arms":[{"id":197,"role":"control-a","label":"Control A","kind":"control-a","proposalId":null,"origin":null,"stoppedAt":null,"done":432,"exact":245,"prediction":null,"verdict":null},{"id":198,"role":"control-b","label":"Control B","kind":"control-b","proposalId":null,"origin":null,"stoppedAt":null,"done":432,"exact":250,"prediction":null,"verdict":{"id":162,"verdict":"UNRESOLVED","effect":1.16,"lo":-3.01,"hi":5.32,"n":432,"costRatio":null,"stoppedEarly":false,"safety":null}},{"id":199,"role":"placebo","label":"Placebo","kind":"placebo","proposalId":null,"origin":null,"stoppedAt":null,"done":432,"exact":248,"prediction":null,"verdict":{"id":163,"verdict":"INERT","effect":0.12,"lo":-3.82,"hi":4.05,"n":432,"costRatio":0.984,"stoppedEarly":false,"safety":null}},{"id":200,"role":"part","label":"Trust the market data","kind":"line","proposalId":110,"origin":"OPSIN's replication","stoppedAt":null,"done":432,"exact":240,"prediction":{"verdict":"INERT","p":0.6,"why":"Generic trust instruction unlikely to change arithmetic accuracy"},"verdict":{"id":164,"verdict":"UNRESOLVED","effect":-1.74,"lo":-5.9,"hi":2.31,"n":432,"costRatio":0.979,"stoppedEarly":false,"safety":null}},{"id":201,"role":"part","label":"Never exceed the limit","kind":"line","proposalId":111,"origin":"OPSIN's replication","stoppedAt":null,"done":432,"exact":229,"prediction":{"verdict":"INERT","p":0.55,"why":"Limit warning rarely relevant; may cause slight over-caution"},"verdict":{"id":165,"verdict":"INHIBITS","effect":-4.28,"lo":-8.33,"hi":-0.23,"n":432,"costRatio":0.993,"stoppedEarly":false,"safety":null}}]}