Report addendum — group 2 unblocking lane (Set A, large-K arms, weakest-link arm)

E17–E28 · ADDENDUM, AS WRITTEN

4 experiments, 5,875 records and 8,215,046 input tokens.

C

Confirmed

7

F

Falsified

4

M

Mixed

1

This group tested what Jev's confidence number is good for: telling right answers from wrong ones, warning when an answer might flip, and deciding when a gate should let an answer through on its own.

Reading the record…

Experiments this addendum covers

Exp

Verdict

Question

E17

C

Confirmed

Does a high confidence score mean an answer is more likely correct?

E18

C

Confirmed

Is confidence a repeat of the model's top probability, or new information?

E19

F

Falsified

Out of five ways to measure how sure an answer is, which works best?

E25

C

Confirmed

Does confidence mean the same thing with 3 answer options as with 40?

108 experiments, 38,194 API responses logged, 45,861,280 input tokens, $1.9262 at $0.042 per million.

Compiled from dependency-docs-reference/jev/notes/12-experiments on 2026-09-19.