Layer 6 · Neuron 3786 · boosted

L6N3786+1

Word quirks:

“ here”“!”“ my”“ now”“ literally”“!"”“ reason”“ myself”

💤 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 Qwen, a large language model developed by Alibaba. My mission is to assist users in understanding complex concepts, answering questions, or generating creative content by leveraging my advanced language capabilities. How can I help you with that? 😊

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. L6N7980+1💤	0.85	“ my”“ now”“ myself”
2. L5N2601+1🕒	0.79	“!”“ here”“ now”
3. L6N2003-1🕒	0.77	“ my”“ myself”“ ‘”
4. L12N3586+1💤	0.75	“!”“ this”“ my”
5. L10N536+1💤	0.74	“ here”“ myself”“ now”
6. L12N4453-1🕒	0.74	“ myself”“ here”“ my”
7. L3N6946+1🕒	0.74	“!"”“ my”“:"”
8. L4N4895-1🕒	0.74	“!”“ now”“ here”
9. L4N7814+1🕒	0.73	“!”“ now”“!"”
10. L9N5770-1💤	0.73	“ my”“ myself”“ ‘”

Shadow specimen, apart from the ten: L6N3786-1🕒, 0.14.

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 6, neuron 3786, 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
L6N3786+1 t0-v0.1	0.928	0.62	1.116	1.51	Top boosted tokens ↑
L6N3786-1🕒 shadow specimen t0-v0.1	1.603	0.51	1.976	1.80	“<\|im_end\|>”“ exhaustive”“ completeness”“⏎”“ complete”“...\”“ logic”“ logical”

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.