Layer 11 · Neuron 6741 · suppressed

L11N6741-1

Word quirks:

“ specific”“ **”“ **"”“ particular”“ specifically”“ Specific”“ **“”“ generally”

🕒 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 Qwen (通义千问).

My mission is to provide high-quality assistance in the following areas:

Core Capabilities: I am trained on Alibaba's internal data, which includes information from various sources such as:
Tongyi Lab (通义实验室)
Alibaba Cloud (阿里云)
Ant Group (蚂蚁集团)
Huawei Cloud (华为云)
JD.com (京东)
Suning (苏宁)

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. L11N7440+1💤	0.94	“ ”“ “”“ specific”
2. L11N8216+1🕒	0.93	“ **"”“ specific”“ specifically”
3. L13N4192+1💤	0.93	“ specifically”“ specific”“ **”
4. L11N4682+1🕒	0.91	“ "”“ ”“ specific”
5. L13N4657-1💤	0.89	“ specific”“ specifically”“ **”
6. L8N7896-1💤	0.87	“ specific”“ **”“ typically”
7. L9N7096-1🕒	0.87	“ specific”“ general”“ special”
8. L12N6874+1🕒	0.87	“ specific”“ particular”“ **”
9. L12N2080+1💤	0.87	“ “”“ specific”“ exact”
10. L8N6618-1💤	0.86	“ specific”“ specifically”“ **“”

Shadow specimen, apart from the ten: L11N6741+1🕒, 0.20.

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 6741, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, top-1 agreement 0.62) 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
L11N6741+1🕒 shadow specimen t0-v0.1	1.115	0.57	1.405	1.47	“ now”“ appropriate”“ ⏎⏎”“ reason”“ ‘”“ ”“ future”“ suitable”
L11N6741-1 t0-v0.1	0.888	0.62	1.118	1.47	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.