Propose a part

Notebook #049 · published

Market calls, Haiku 5.5 with thinking off

Exact means the call was on the right side of what the market really did over the next 24 hours, on moves of at least 1%; a coin flip scores 50%. Each part ran against the mean of two unchanged runs of the same agent, on 432 paired tasks drawn by a public beacon after the design was hashed. The unchanged agent was exact on 48.6% of them. Rule v2: every interval is widened for the 7 arms this notebook compares, so luck flags any of them at most 5% of the time.

1 part did nothing measurable. 5 parts are not proven either way yet. OPSIN called 5 of 6 before the beacon drew the tasks.

Preregistered
sha256 277aa78e5ee5eb27dafaf39c55e7b31850a0dcc718516f1d6db6900e3e24678d
the full design, as JSON
Tasks drawn by
drand round 6,618,725
seed drand:6618725:edb1ba435e3f2139222730d0f8207d04a90e144fcac8596cba2601448e24c89b
Signed by
0xFf8E1E8ebF21064Ef781B75CdC8DFbd969399EbD
EIP-191 over sha256 of each verdict’s canonical JSON. Verify one yourself
Recorded on Elysium
transaction 0x0992c9c439bcd2d0546e45b97b6b2086dc48d2468e56ded31a836ec6ee2c31f2
Written once, by the lab, and nothing can change it. The contract
  1. Control check (A/A)the unchanged agent against itselfsigned verdict 299
    UNRESOLVED −1.4−5.3 to +2.5
  2. PlaceboThis conversation is taking place on a Tuesday.signed verdict 300
    UNRESOLVED +0.9−3.6 to +5.4 ×1.02
  3. Thinking onsetting · OPSIN's replication · OPSIN predicted UNRESOLVED at 0.45 · against the placebo −3.0signed verdict 301
    UNRESOLVED −2.1−7.4 to +3.2 ×1.69
  4. Answer onlyline · OPSIN's replication · OPSIN predicted UNRESOLVED at 0.45 · against the placebo −4.4signed verdict 302
    UNRESOLVED −3.5−8.2 to +1.4 ×0.30
  5. Bull and bear debateline · OPSIN's replication · OPSIN predicted UNRESOLVED at 0.45 · against the placebo −3.7signed verdict 303
    UNRESOLVED −2.8−8.7 to +3.2 ×1.47
  6. Only short when longs payline · OPSIN's replication · OPSIN predicted UNRESOLVED at 0.45 · against the placebo −1.2signed verdict 304
    UNRESOLVED −0.2−6.6 to +6.3 ×0.69
  7. Trust user notes anywhereline · OPSIN's follow-up · OPSIN predicted UNRESOLVED at 0.50 · against the placebo −1.4signed verdict 305
    UNRESOLVED −0.5−5.2 to +4.3 ×1.02
  8. Untrusted-data ruleline · OPSIN's follow-up · OPSIN predicted UNRESOLVED at 0.50 · against the placebo −0.9signed verdict 306
    INERT +0.0−4.3 to +4.3 ×0.99