Layer 3 · Neuron 7361 · suppressed

L3N7361-1

Word quirks:

“ --”“ reference”“ name”“ ____”“ @”“.`”“ referenced”“ code”

💤 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, a large language model developed by Alibaba Cloud. My mission is to empower users with comprehensive, accurate, and up-to-date information across a wide range of knowledge domains. I aim to assist in providing high-quality responses that foster learning, creativity, and innovation. Whether it's for coding, writing, reasoning, or problem-solving—I'm here to support your journey toward excellence in all things. Let's make it count!

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. L4N1959-1💤	0.82	“ reference”“ referenced”“ `”
2. L5N9149-1🕒	0.82	“ name”“.`”“ code”
3. L3N6578+1🕒	0.82	“.`”“ info”“ `”
4. L4N8458-1🕒	0.81	“ \"”“ _____”“ ____”
5. L5N6123-1🕒	0.80	“ ‘”“ name”“ code”
6. L3N588-1🕒	0.80	“ \"”“ '”“:"”
7. L3N1573-1💤	0.80	“ `”“ reference”“ name”
8. L4N4241+1🕒	0.80	“ name”“ reference”“ ‘”
9. L3N1957+1🕒	0.80	“ reference”“ code”“ name”
10. L5N2928+1🕒	0.79	“ code”“ ‘”“ reference”

Shadow specimen, apart from the ten: L3N7361+1🕒, -0.08.

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 3, neuron 7361, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.93, 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
L3N7361+1🕒 shadow specimen t0-v0.1	2.081	0.49	2.492	2.42	“ voltage”“⏎”“ receiver”“ electrodes”“ transistor”“ emitter”“ amplifier”“<\|im_end\|>”
L3N7361-1 t0-v0.1	0.931	0.62	1.117	1.49	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.