Layer 18 · Neuron 5407 · boosted

L18N5407+1

Word quirks:

“ theirs”“ their”“ his”“ they”“ ones”“—they”“让他们”“ them”

💤 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 Tencent. My primary purpose is to assist users in completing tasks efficiently and effectively. As an advanced AI, I aim to provide accurate information, solve problems, and support their needs across various fields.

Key Features:

Accurate & Reliable: Designed for precision in understanding and responding to queries.
Efficient
User-Friendly
Collaborative

Let me help you with what you need! 😊

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. L14N4679-1💤	0.93	“ they”“ them”“ it”
2. L17N1460+1🕒	0.92	“ themselves”“—they”“ ourselves”
3. L16N4807-1🕒	0.92	“ themselves”“ they”“ their”
4. L13N6194-1🕒	0.92	“ their”“ theirs”“ maybe”
5. L18N7820-1🕒	0.91	“ their”“ they”“Their”
6. L12N6872-1🕒	0.91	“ their”“ they”“ They”
7. L15N7526-1🕒	0.91	“ theirs”“ their”“他们的”
8. L13N8568-1💤	0.90	“ their”“ themselves”“ they”
9. L16N4807+1💤	0.90	“ their”“ they”“他们的”
10. L19N5415+1🕒	0.90	“ them”“ theirs”“ ones”

Shadow specimen, apart from the ten: L18N5407-1💤, 0.26.

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 18, neuron 5407, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.56, top-1 agreement 0.74) 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
L18N5407+1 t0-v0.1	0.562	0.74	0.646	1.08	Top boosted tokens ↑
L18N5407-1💤 shadow specimen t0-v0.1	0.559	0.72	0.693	1.19	“ wherein”“ oneself”“ constitutes”“ itself”“ subsequently”“ constitute”“ methodologies”“ quant”

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.