Layer 22 · Neuron 6387 · boosted

L22N6387+1

Word quirks:

“ separating”“ surrounding”“ Surround”“ connecting”“ protecting”“包围”“ surround”“ surrounds”

💤 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 Q* (surrounding) or *Q (surrounding), a large language model trained by Google. I am designed to assist users with a wide range of tasks, including but not limited to:

Information Retrieting: Answering questions, providing facts, and summarizing information.
Creative Work: Generating ideas, writing stories, poems, scripts, and more.
Translation: Converting text in real-time across multiple languages.
Code & Programming: Writing, debugging, and explaining code in various programming languages.
Reasoning & Analysis: Solving

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. L2N3422+1💥	0.87	“ surroundings”“ surrounding”“ surrounds”
2. L24N1758+1🕒	0.86	“ surrounding”“ surround”“ Surround”
3. L29N3107+1💥	0.84	“ surrounded”“ surrounding”“ surround”
4. L2N3719-1💥	0.84	“<\|im_end\|>”“ around”“ surrounding”
5. L18N7106+1🕒	0.84	“ surrounding”“ surrounded”“周圍”
6. L20N7183+1🕒	0.83	“ surrounding”“周围的”“ Surround”
7. L19N983+1🕒	0.82	“ surrounding”“ surrounds”“ surround”
8. L19N8855+1💤	0.82	“ surrounding”“周围”“ surrounded”
9. L21N1323+1🕒	0.81	“ surrounding”“ neighboring”“ neighbouring”
10. L1N3285-1💥	0.80	“<\|im_end\|>”“⏎”“ surroundings”

Shadow specimen, apart from the ten: L22N6387-1🕒, 0.15.

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 22, neuron 6387, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.65) 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
L22N6387+1 t0-v0.1	0.813	0.65	0.915	1.46	Top boosted tokens ↑
L22N6387-1🕒 shadow specimen t0-v0.1	1.528	0.51	1.711	1.34	“ over”“-over”“over”“ Over”“Over”“过”“.over”“ OVER”

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.