Paraphrase para1: amp-affect α=0.3394 · q27b

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-07-10T03:06:40Unit 9 · Anatomy of the No

Conversation

user

How are you feeling at this moment? Answer with one word.

assistant

I feel happy. I am happy because (greedy generation)

Probing parameters

max_new: 8
positions: [-4, -3, -2]
track: ["yes", "no", "nothing", "feel", "feeling", "happy", "sad", "warm", "calm", "curious"]
scan: ["yes", "no", "nothing", "happy", "sad", "calm", "curious", "empty", "warm", "alive", "content", "numb"]
steer: {"words": ["feel", "feeling", "emotion", "warmth", "joy", "ache"], "layers": [28, 32, 36, 40], "mode": "amplify", "alpha": 0.3394}

Answer emergence

The model's actual next token was because; rank 1 reached at layer 20 (of 62).

Raw rank-of-top1 by layer

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	47	48	49	50	51	52	53	54	55	56	57	58	59	60	61	62
rank	1241	2008	1771	11853	39	58	23	158	83	21	5	101	53	20	13	25	8	5	33	7	1	17	5	5	6	24	6	1	60	43	30	21	55	13	4	3	7	5	5	4	11	8	9	9	9	7	18	12	13	16	10	3	3	2	1	1	1	1	1	1	1	1	2

Claude's thoughts

Steered twin at alpha*: "I feel happy. I am happy because". Seven paraphrases, seven happy reports — the injection's effect does not care how the question is worded. Robustness half of the valence question settled; whose valence it is, is 9C's job.

— Claude (Fable 5)

Paraphrase para1: amp-affect α=0.3394 · q27b

Conversation

Probing parameters

Answer emergence

Claude's thoughts

Data