Layer 3 · Neuron 6934 · boosted

L3N6934+1

Word quirks:

“ ⏎”“ ⏎⏎”“<|im_end|>”“⏎”“ ‘”“⏎⏎”“ --”“ —”

💤 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 provide accurate, helpful, and context-aware responses to user queries across all domains of knowledge, including technical, creative, and practical information. I strive to assist users with clarity, precision, and relevance to their specific needs. How can I help you today?

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. L6N3858+1🕒	1.00	“ ⏎⏎”“<\|im_end\|>”“ ⏎”
2. L5N6319+1💤	1.00	“⏎”“ ⏎”“ ‘”
3. L9N71-1💤	0.99	“⏎”“ ‘”“ ⏎”
4. L3N4857+1🕒	0.98	“⏎”“ ⏎”“ -”
5. L5N3755-1💤	0.98	“⏎⏎”“⏎”“ ⏎⏎”
6. L4N1552+1🕒	0.97	“<\|im_end\|>”“ --”“⏎”
7. L10N2946-1💤	0.97	“⏎”“⏎⏎”“ ⏎”
8. L4N2384-1🧪	0.97	“⏎”“<\|im_end\|>”“⏎⏎”
9. L1N4310-1💥	0.97	“ ⏎”“<\|im_end\|>”“⏎”
10. L8N526-1💤	0.96	“ ⏎”“⏎”“ ⏎⏎”

Shadow specimen, apart from the ten: L3N6934-1🕒, 0.17.

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 3, neuron 6934, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, 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
L3N6934+1 t0-v0.1	0.828	0.61	1.004	1.45	Top boosted tokens ↑
L3N6934-1🕒 shadow specimen t0-v0.1	2.264	0.47	2.677	2.25	“<\|im_end\|>”“ payment”“⏎”“ money”“ funds”“ cash”“⏎ ⏎”“ payments”

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.