Today
-
Card 1.08b: pre-registered metric (all 360 items) first
-
Card 1.08b: held-out v2 result (no measurable reasoning gain over the control)
-
Card 1.08b erratum: panel was computed before the verdict (reported separately)
Yesterday
-
Card 1.08b: standard benchmark panel (regression in both arms vs pre-registered guards)
-
Card 1.08b: evaluation result (inconclusive on reasoning, measured Polish retention)
-
Model card: describe the trained experimental run 1.08b (reasoning arm) and its control
-
Slayer1B 1.08b: model card (in preparation)
