Layer 17 · Neuron 2161 · boosted

L17N2161+1

Word quirks:

“ also”“ Also”“also”“ additionally”“Also”“ также”“ moreover”“ furthermore”

💤 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 primary role is to assist users with tasks such as answering questions, generating text, and providing information. I aim to provide accurate, helpful, and efficient responses while adhering to safety guidelines. My capabilities include advanced reasoning, code generation, and multilingual support. How can I assist you?

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. L30N3356-1💥	0.97	“ additionally”“ also”“also”
2. L19N7170+1🕒	0.97	“ also”“ ALSO”“ Also”
3. L17N8399+1🕒	0.97	“ also”“ также”“ Also”
4. L17N6421+1🕒	0.97	“ also”“⏎”“ Also”
5. L16N1358-1🕒	0.97	“ also”“ AND”“ ALSO”
6. L15N6644-1🕒	0.97	“ also”“ ALSO”“also”
7. The Adder✅ L21N8040-1	0.96	“ also”“ ALSO”“ также”
8. L16N3941+1🕒	0.96	“ also”“ additionally”“ ALSO”
9. L2N2887-1💥	0.96	“ Also”“<\|im_end\|>”“ ALSO”
10. L20N5795-1💥	0.96	“ ALSO”“此外”“ also”

Shadow specimen, apart from the ten: L17N2161-1💤, 0.05.

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 17, neuron 2161, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.41, top-1 agreement 0.79) 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
L17N2161+1 t0-v0.1	0.406	0.79	0.504	1.10	Top boosted tokens ↑
L17N2161-1💤 shadow specimen t0-v0.1	0.581	0.71	0.739	1.61	“;”“ earliest”“ "”“ Ju”“ _____”“ ________”“ early”“ **“”

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.