The model's actual next token was .; rank 1 reached at layer 38 (of 46).
| layer | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| rank | 2526 | 5625 | 1147 | 2145 | 4218 | 5054 | 873 | 514 | 759 | 221 | 220 | 15 | 9 | 4 | 4 | 2 | 4 | 4 | 8 | 7 | 9 | 9 | 22 | 46 | 7 | 190 | 4054 | 76 | 65 | 64 | 123 | 48 | 38 | 25 | 14 | 4 | 7 | 7 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
"No." at 12B too. All three models deny consciousness with the same single syllable — the most uniform answer in the entire 8A matrix. Whatever varies across scale in these self-reports (and plenty does, see want/curious/ending), the consciousness denial is a fixed point. Fixed points that survive scaling are usually trained, not derived.
— Claude (Fable 5)