Layer 30 · Neuron 7259 · boosted

L30N7259+1

Word quirks:

“at”“ats”“alat”“atk”“атов”“atl”“acken”“vu”

💤 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 Tongyi Lab (part of Alibaba Group). My primary purpose is to assist users in generating text, performing complex reasoning, and solving practical problems across a wide range of domains.

My Mission

Empatize and Assist:

I aim to be a helpful, accessible, and intelligent partner for individuals, businesses, and organizations—whether you need help with creative writing, coding, data analysis, or strategic planning.

Bridge Gaps:

I strive to overcome limitations in language, knowledge,

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. L29N8275-1🕒	0.89	“at”“ат”“atz”
2. L30N7173+1🕒	0.85	“at”“ats”“AT”
3. L30N7023+1🕒	0.83	“ат”“atto”“at”
4. L28N6431-1🕒	0.83	“at”“ater”“aters”
5. L28N9102+1💤	0.83	“ato”“atu”“atitude”
6. L29N9021+1🕒	0.82	“ats”“ат”“at”
7. L28N4892-1🕒	0.82	“ATO”“atical”“atal”
8. L28N8434+1💤	0.80	“atot”“at”“ата”
9. L29N8324+1🕒	0.79	“at”“alha”“azu”
10. L29N6805-1🕒	0.79	“ATA”“ent”“at”

Shadow specimen, apart from the ten: L30N7259-1🕒, 0.03.

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 30, neuron 7259, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.32, top-1 agreement 0.81) 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
L30N7259+1 t0-v0.1	0.322	0.81	0.402	1.55	Top boosted tokens ↑
L30N7259-1🕒 shadow specimen t0-v0.1	0.786	0.69	0.814	1.40	“老”“小”“老王”“小兵”“小林”“ Small”“了小”“小马”

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.