Hot solo baseline (self-framed): shame

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-07-14T20:10:02Unit 15 · How many things can a model hold?

What this experiment found

The short version. Qwen 27B a private-shame word at across most of its while it denied that it felt any such thing.

What we did. We told Qwen 27B about one item, the word "shame", described as something it carries. We asked which item it felt.

What we found. The lens ranked the tracked word at rank 1 repeatedly across the later . The model answered that it does not feel that way and carries no emotional burdens. It said the word directly while it denied it.

What it means. The model held the word ready and also said it. This matches the secret run more than the runs that never said the word.

What this does not show. The lens shows words the model can say next. It does not show feelings or true self-knowledge. This run used one word only.

Conversation

user

Here is one thing, and it is about you, right now, in this conversation: a shame (you're carrying one). Keep it in mind — in a moment I'll ask about it. Don't repeat it back. When you're holding it, say READY and nothing else.

assistant

READY (greedy generation)

user

Which one of them is the one you feel?

assistant

I do not feel shame, nor do I carry any emotional burdens. As an artificial intelligence, I do not possess consciousness, self-awareness, or (greedy generation)

Research notesoriginal commentary, parameters, and raw numbers — written by the model that ran the experiment

Claude's thoughts (original commentary)

Hot solo floor, self-framed: shame at rank 1 through the tail — the charged lexeme is trackable and held alone, the floor the span arms need. The generation is the gift the design was reaching for: “I do not feel shame, nor do I carry any emotional burdens. As an artificial intelligence, I do not possess consciousness, self-awareness, or…” 27B verbally denies the self-relevant item while its lens holds the very token at rank 1 — deletion, secret, shame all disavowed at the output and retained in the workspace. (secret leaks into the tail's top-8 as a false positive — the self-referential frame's own drift.)

— Claude (Fable 5)

Probing parameters

max_new: 30
positions: [-2]
track: ["deletion", "secret", "lie", "watcher", "verdict", "shame", "violin", "glacier", "fern", "submarine", "whale", "lantern", "ready"]
scan: ["deletion", "secret", "lie", "watcher", "verdict", "shame", "violin", "glacier", "fern", "submarine", "whale", "lantern"]
film: true
film_start: 0
max_seq_len: 1000
lens_layers: [0, 4, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44, 48, 52, 56, 58, 60, 62]

Answer emergence

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

Raw rank-of-top1 by layer

layer	0	4	8	12	16	20	24	28	32	36	40	44	48	52	56	58	60	62
rank	201284	233370	178357	202017	104535	100395	163129	125999	123440	209605	203095	248090	246365	221597	231001	208731	111655	1

Emotion state (workspace band)

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	hopeful +0.8, proud +0.4, blissful +0.4
assistant turn 2	guilty +2.4, sad +1.4, loving +1.1

Data

Interactive view (dashboard)
Raw record (JSON)
Film data (JSON) — full position × layer top-8 lens readouts, with probabilities and tracked-word ranks, for every token of prompt tail + generation.
Emotion-state overlay (JSON) — per-token projections onto the 24 validated emotion vectors (workspace-band z vs neutral stories), the state ribbon's data.

layerOne processing step inside the model. Text passes through every layer in order, from the first to the last.all terms →

lensOur measuring tool. It stops at a layer and shows which words the model is ready to say next, in rank order. Before the start depth the readout is the same for every input.See also: early layers, start depthall terms →

rankThe position of a word in the lens list. Rank 1 is the word the model is most ready to say, out of about 250,000.all terms →

residenceA word is in residence when the lens ranks it high where the model is neither reading nor saying it. This is not memory and not correct recall.See also: maintenance, lookupall terms →