Layer 7 · Neuron 6250 · boosted

L7N6250+1

Word quirks:

“⏎”“ ⏎”“ ______”“ ________”“ ____”“ __”“ _____”“ crafted”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, a large language model developed by the Tongyi Lab team at Alibaba Cloud. My mission is to provide users with high-quality, accurate, and helpful information that can assist in solving complex problems and advancing scientific research. I am committed to fostering innovation and collaboration across industries, and I believe in the power of technology to transform the world. Thank you for your interest in this project!

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. L14N1095+1🕒	0.97	“⏎”“ ⏎”“ ____”
2. L10N3533+1🕒	0.97	“ ____”“ ⏎”“ ______”
3. L7N2559-1🕒	0.97	“ __”“ ____”“ _____”
4. L16N830+1🕒	0.97	“⏎”“ ⏎”“ ____”
5. L8N192+1💤	0.97	“ ____”“⏎”“ ______”
6. L8N847+1💤	0.96	“⏎”“ ⏎”“ ______”
7. L7N7954+1🕒	0.96	“ ⏎”“⏎”“ ____”
8. L7N3865-1💤	0.96	“⏎”“ ⏎”“ ________”
9. L12N5245+1🕒	0.96	“ ⏎”“ ____”“ ______”
10. L12N1535-1🕒	0.96	“⏎”“ ⏎”“ ⏎”

Shadow specimen, apart from the ten: L7N6250-1🕒, 0.84.

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 7, neuron 6250, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.68) 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
L7N6250+1 t0-v0.1	0.798	0.68	0.998	1.35	Top boosted tokens ↑
L7N6250-1🕒 shadow specimen t0-v0.1	1.159	0.59	1.452	1.60	“⏎”“<\|im_end\|>”“ ⏎”“ _____”“ ____”“?\”“ ⏎”“ ______”

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.