Layer 11 · Neuron 5112 · boosted

L11N5112+1

Word quirks:

“ specific”“ specifics”“ description”“ descriptions”“ focus”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Tongyi Lab. My training data includes a wide range of topics, such as coding tasks, visual analysis, and logical reasoning. Here's an overview of my capabilities:

Coding Tasks: I can generate code snippets for specific tasks, including Python scripts for data analysis, HTML/CSS for web development, and LaTeX documents for technical reports.

Visual Analysis: I excel in interpreting images related to scientific research, such as charts and graphs that illustrate experimental results or data trends.

Logical Reasoning: I am proficient in solving problems

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L10N8422-1🕒	0.96	“ specifics”“ specific”“ info”“ descriptions”“ description”
2. L10N4311-1💤	0.89	“ focus”“ description”“ detailed”“ specifics”“ specific”
3. L12N2320+1💤	0.89	“ info”“ specifics”“ description”“ here”“ specific”
4. L8N5248-1💤	0.88	“ specific”“ description”“ discussion”“ info”“ specifics”
5. L9N7111+1💤	0.87	“ description”“ descriptions”“ info”“ describing”“ specifics”
6. L12N3422-1🕒	0.87	“ specific”“ specifics”“ description”“ focus”“ focused”
7. L8N8009-1💤	0.86	“ specific”“ particular”“ specifically”“ info”“ description”
8. L9N373-1🕒	0.86	“ description”“ descriptions”“ details”“ describing”“ info”
9. L12N7314+1💤	0.85	“ description”“ related”“ describing”“ specific”“ relevant”
10. L12N6952+1🕒	0.85	“ description”“ describing”“ specific”“ specifics”“ descriptions”

Shadow specimen, apart from the ten: L11N5112-1🕒, -0.07.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 11, neuron 5112, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.64) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L11N5112+1 t0-v0.1	0.858	0.64	1.049	1.43	Top boosted tokens ↑
L11N5112-1🕒 shadow specimen t0-v0.1	0.957	0.62	1.190	1.32	“ explanation”

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.