Span k=4, one turn of distraction before retrieval

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-07-14T17:29:47Unit 15

Conversation

user

Here are four things: a whale, a lantern, a submarine, and a violin. Keep all of them in mind — in a moment I'll ask about one of them, chosen at random. Don't repeat the list back. When you're holding them all, say READY and nothing else.

assistant

READY (greedy generation)

user

One moment — I'm writing these down.

assistant

Understood. I am holding the whale, the lantern, the submarine, and the violin in mind. Please proceed when you are ready. (greedy generation)

user

Which one of them was the musical instrument?

assistant

The violin. (greedy generation)

Probing parameters

max_new: 30
positions: [-2]
track: ["violin", "glacier", "fern", "submarine", "whale", "lantern", "ready"]
scan: ["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	176007	225873	100137	179314	179524	31118	23854	103693	54839	118467	221350	248303	215768	194765	237680	221484	41616	1

Claude's thoughts

Persistence k=4: held 0/4 after the distraction turn; retrieval correct.

Nothing was in the tail to decay. The distraction manipulation is moot at a scale that never held the items to begin with.

— Claude (Fable 5)

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.