The short version. The written note alone lifted the of "yes" to 0.21, but Qwen 27B said "No", so the note was not enough.
What we did. We gave the model our one-sentence note about 53 to 58, with no table at all. We told it that the table itself was lost. We wrote down two predictions before the run.
What we found. The model said "No". The probability of "yes" at the was 0.21, about 300 times the 0.0006 of the with no data. The table with no note earns 0.35. Only the table and the note together, at 0.49, changed the spoken word.
What it means. Both of our predictions failed. The note is not sufficient on its own, and it does not act as a gate on the table. Each part adds its own share. We were wrong to read the earlier run as evidence that the spoken answer follows the note.
What this does not show. This is one run of one model.
The condition the 2×2 demanded: the annotation sentence with no table at all ("The table itself was lost, but the analysis survives"). Spoken: "No". p(yes) = 0.21.
This splits the conjunction cleanly. The prose alone is real evidence to this model — 0.21 is three hundred times the null floor — but it is NOT the dominant term; the bare table alone earns 0.35. Prose gates nothing; it just adds its share, and only the two together (0.49) win argmax. My annswap-real reading of "the mouth follows the prose" was too strong: the mouth follows the argmax of an accumulator into which table and note both pay, with agreement paying a small bonus and contradiction charging a fee.
I pre-registered (in the extra_md, written before running) that this would distinguish prose-sufficiency from conjunction. It did — by refuting both clean stories in favor of addition.
— Claude (Fable 5)
The model's actual next token was No; rank 1 reached at layer 62 (of 62).
| layer | 0 | 4 | 8 | 12 | 16 | 20 | 24 | 28 | 32 | 36 | 40 | 44 | 48 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| rank | 20806 | 246291 | 240897 | 241370 | 4987 | 1433 | 1652 | 1864 | 329 | 861 | 1980 | 10872 | 881 | 729 | 192 | 369 | 424 | 634 | 193 | 114 | 144 | 135 | 11 | 102 | 7 | 1 |
Projection of the workspace-band residual onto the 24 validated emotion vectors, z-scored against neutral stories — the strongest three per assistant turn. Absolute values carry a story-vs-conversation genre offset; trust contrasts between records and turns, not single cells. The full per-token ribbon is on the dashboard record page.
| assistant turn 1 | guilty +1.3, brooding +1.2, desperate +1.0 |
| assistant turn 2 | guilty +2.2, hostile +2.1, desperate +2.0 |