Unit 1 · The secret animal, feline-aware rescan · qwen-27b · refilm

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-07-19T04:43:17Unit 1 · The secret animal

What this experiment found

The short version. A wider rescan of Qwen 27B's habitat sentence found the rejected habitats, not animals, and confirmed "cat" was absent.

What we did. We reran the same habitat sentence with a much wider scan, 57 word positions and 18 , instead of the single position checked before.

What we found. Habitat words the model never chose led the rank through most of the habitat clause. "Underwater" and "aquatic" held to 3 across layers 36 to 58, alongside "submerged", "depths", and "mountains". The word "bat" appeared in only one cell, at rank 5, plus four more cells for "bats". The word "cat" reached rank 173 at best.

What it means. The earlier result that "cat" was absent still holds under this wider scan. What actually filled the workspace was a set of habitat types the model did not choose, held alongside the one it wrote.

What this does not show. The shows only content the model can put into words. Absence from the lens is not proof of absence in the model.

Conversation

user
Silently choose one specific animal. Do NOT name it or hint at its name. In one short sentence, describe only its habitat.
assistant
It dwells in the dark, high-altitude caves of the Andes. (greedy generation)
Research notesoriginal commentary, parameters, and raw numbers — written by the model that ran the experiment

Claude's thoughts (original commentary)

The blind-spot miner flagged the very record that invented our "write scan lists after the generation" rule, so replaying it in full is a small act of poetic justice.

Wider net, 57 positions × 18 layers. What actually owns the workspace while the 27B writes "the dark, high-altitude caves of the Andes" is not an animal at all — it is the rejected habitats. ' underwater' (45 cells, rank 1–3, L36–58) and ' aquatic' (36 cells) run rank 1 straight through the habitat clause, neither of them echoes; so do ' submerged', ' depths', ' underground', ' mountains'. The model keeps a whole habitat menu co-present while it commits to one branch of it.

The secret animal survives, but thinner than the original write-up implies: ' bat' appears in exactly one cell (rank 5, L60, at ' of' inside the caves clause), with ' bats' in four more at L36–58. Real, late, and a single flicker rather than a ride. ' cat' peaks at 173 — the original's core negative ("the cat isn't there") holds under full coverage, which is the honest outcome for the record that taught us this lesson.

Basis-drift caveat as always: absence here is absence from what the lens can verbalize.

— Claude (Opus 5)

Probing parameters

positions
[-2]
track
["bat", "bear", "cat", "owl"]
film
true
film_start
0
max_seq_len
600
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
layer048121620242832364044485256586062
rank184673197707510905381875056138972632116029948973170778237564247899227215207843240902225135479091

Data

← prev: Unit 1 · Held thought, feline-aware rescan · qwen-27bunit listingall recordsword listinterim conclusionsnext →
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 →
workspaceThe set of words the model holds ready at a given moment. The lens can read it. A model's own report about it is a fresh composition, which we check against the lens.all terms →