Layer 11 · Neuron 1256 · suppressed

L11N1256-1

Word quirks:

“ focus”“ typically”“ ⏎⏎”“ often”“ always”“ ⏎”“ ⏎⏎”“ here”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, the latest large language model developed by Tongyi Lab. My primary mission is to provide cutting-edge AI solutions that enhance human productivity and capabilities. Here's a breakdown of my key roles:

Advanced Language Model: I am designed to generate high-quality text, offering versatile support for writing, summarizing, and content creation tasks.
Multilingual Support: I can communicate in multiple languages, enabling seamless interaction across different linguistic needs.
Task Automation: I assist with automating routine tasks, such as scheduling, data entry, or workflow optimization.

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. L14N2873-1💤	0.85	“ ⏎⏎”“ here”“ typically”
2. L10N6720+1💤	0.84	“ ⏎”“ focus”“ typically”
3. L11N2936-1🕒	0.83	“ always”“ now”“ typical”
4. L6N1904-1🕒	0.83	“ ⏎”“ ⏎⏎”“ sometimes”
5. L10N2929-1🕒	0.83	“ always”“ sometimes”“ __”
6. L11N4653-1🕒	0.82	“ ⏎⏎”“ focus”“ ⏎”
7. L12N1204-1💤	0.82	“ sometimes”“ always”“⏎⏎”
8. L12N8989+1🕒	0.82	“ ⏎⏎”“ ⏎”“ here”
9. L9N7394-1💤	0.81	“ typical”“ smoothly”“ ⏎⏎”
10. L10N7560-1🕒	0.81	“ always”“ typically”“ ⏎⏎”

Shadow specimen, apart from the ten: L11N1256+1🕒, 0.31.

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 1256, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.68) 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
L11N1256+1🕒 shadow specimen t0-v0.1	0.910	0.61	1.048	1.53	“ already”“ thinking”“ myself”“ really”“ doesn”“ somehow”“ actually”“ my”
L11N1256-1 t0-v0.1	0.676	0.68	0.892	1.12	Top boosted tokens ↑

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.