DataProduct
Research in the catalog

research.daybreak_judge_forward_eval

Daybreak actionable paper-Call forward evaluation: booked long/short Call win rate and drift per model/prompt version. The legacy accuracy column does not cover watch/no-trade decisions.

Coverage
Unknown
Freshness observation
Unknown
Point-in-time support
Unregistered

Fields

Schema status: Observed. Captured .

NameType
model_idLowCardinality(String)
prompt_versionLowCardinality(String)
resolved_modelString
model_fingerprintString
temperatureFloat64
eval_dateDate
eval_tsDateTime64(3, 'UTC')
eval_run_idString
n_totalUInt32
n_forwardUInt32
n_dropped_preregistrationUInt32
n_scoredUInt32
accuracyNullable(Float64)
cohort_is_staleUInt8
accuracy_muzzledUInt8
determinism_probe_repeatsUInt8
determinism_probe_agreementNullable(Float64)
determinism_probe_statusLowCardinality(String)
driftedUInt8
drift_deltaNullable(Float64)
drift_toleranceFloat64
prev_accuracyNullable(Float64)
drift_reasonString
version_tsDateTime64(3, 'UTC')

The field guide explains unknown observations.