Layer 19 · Neuron 4642 · boosted

L19N4642+1

Word quirks:

“ starting”“starting”“ beginning”“ Starting”“ starts”“ started”“ start”“Starting”

💤 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 Qwen3.8, a large language model developed by Qwen. My primary mission is to provide high-quality, accurate, and diverse information, assist with complex tasks, and facilitate meaningful interactions. I aim to support users in everything from creative writing and logical reasoning to technical problem-solving and multilingual communication.

Key aspects of my mission include:

Accurate Information: Providing reliable data and knowledge across various domains.
Task Assistance: Helping with intricate tasks like code generation, data analysis, and multi-modal content creation.
Efficient Interaction: Enabling seamless conversations, with capabilities ranging

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. L24N5450+1🕒	0.99	“ starting”“ beginning”“ starts”
2. L16N6143-1💤	0.99	“ starting”“ starts”“ start”
3. L19N9194-1💤	0.98	“ starting”“ beginning”“ start”
4. L20N3533+1🕒	0.98	“ starting”“开始”“ start”
5. L21N7539-1🕒	0.98	“ starting”“starting”“起始”
6. L19N1736+1🕒	0.98	“ starting”“ Starting”“起始”
7. L23N5907-1🕒	0.98	“ starting”“ Starting”“starting”
8. L24N6368-1🕒	0.98	“ starting”“starting”“ Starting”
9. L26N3031+1🕒	0.97	“ starting”“ Starting”“ Start”
10. L18N5724-1💤	0.97	“ started”“ start”“ starts”

Shadow specimen, apart from the ten: L19N4642-1💤, 0.16.

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 19, neuron 4642, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.46, top-1 agreement 0.72) 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
L19N4642+1 t0-v0.1	0.460	0.72	0.580	1.19	Top boosted tokens ↑
L19N4642-1💤 shadow specimen t0-v0.1	0.586	0.66	0.683	1.21	“ seeing”“ AFTER”“ see”“ THEN”“ beyond”“ outer”“ basic”“那边”

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.