Layer 11 · Neuron 6236 · suppressed

L11N6236-1

Word quirks:

“ similar”“ similarly”“ analogous”“ exempl”“ description”“ typical”“ analogy”“ same”

💤 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-scale language model developed by the Tongyi Lab. My mission is to provide high-quality assistance in terms of:

Functionality: Delivering clear, accurate, and contextually appropriate responses to user queries.
Style: Maintaining a natural, professional, and engaging tone.
Structure: Organizing information logically and coherently.
Values: Adhering to ethical guidelines and societal norms reflected in my training data.

In summary, I am designed to be a reliable and helpful resource for users across various domains.

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. L14N531-1🕒	0.93	“ similar”“ similarly”“ analogous”
2. L13N6553-1💤	0.91	“⏎”“ similar”“ analogous”
3. L10N4674+1🕒	0.90	“ similar”“ similarly”“ comparable”
4. L14N7492-1💤	0.90	“ similar”“ analogous”“ similarly”
5. L17N3633+1💤	0.90	“ similar”“ comparable”“ above”
6. L10N6953-1🕒	0.90	“ similar”“ analogy”“ similarly”
7. L15N9018+1💤	0.90	“ similar”“ similarly”“ comparable”
8. L14N4270-1💤	0.90	“ analogous”“ similar”“ similarly”
9. L13N3622+1🕒	0.90	“ similar”“ analogous”“ analogy”
10. L13N3497+1🕒	0.89	“ similar”“ similarly”“ analogy”

Shadow specimen, apart from the ten: L11N6236+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 11, neuron 6236, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, top-1 agreement 0.61) 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
L11N6236+1🕒 shadow specimen t0-v0.1	0.983	0.64	1.151	1.27	“!””“!”“ “”“ ‘”“!”.”“!\”“!’”“ ””
L11N6236-1 t0-v0.1	0.769	0.61	1.038	1.30	Top boosted tokens ↑

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.