Layer 14 · Neuron 7155 · suppressed

L14N7155-1

Word quirks:

“ thus”“ clar”“ clarification”“ here”“ unclear”“ clarify”“ clarified”“ focus”

💤 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 Alibaba Cloud. My mission is to assist users with diverse tasks, from answering questions to generating creative content. I aim to provide helpful, accurate information while adhering to ethical guidelines. If you have any questions or need assistance, feel free to ask! 😊

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. L11N2699-1🕒	0.87	“ clarified”“ clarification”“ clar”
2. L11N4849+1🕒	0.87	“ clarification”“ clar”“ clarified”
3. L6N8769-1🕒	0.86	“ clarification”“ clarity”“ clar”
4. L11N2007+1🕒	0.86	“ clarification”“ interpretation”“ clarified”
5. L9N5434-1🕒	0.85	“ clarified”“ clarification”“ clar”
6. L13N5787+1💤	0.85	“ clear”“ clearer”“ clarified”
7. L14N3040+1💤	0.84	“ clarified”“ clearer”“ clar”
8. L17N445+1💤	0.84	“ clarification”“ clar”“ noting”
9. L3N7666-1💥	0.84	“ clarification”“<\|im_end\|>”“ clarify”
10. L11N4637+1💤	0.84	“ here”“ clarification”“ clearer”

Shadow specimen, apart from the ten: L14N7155+1💤, 0.16.

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 14, neuron 7155, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, top-1 agreement 0.66) 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
L14N7155+1💤 shadow specimen t0-v0.1	0.950	0.62	1.101	1.64	“ also”“ similar”“ similarly”“ same”“ basically”“ likewise”“ ALSO”“ additionally”
L14N7155-1 t0-v0.1	0.819	0.66	1.061	1.28	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.