Layer 6 · Neuron 7501 · boosted

L6N7501+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 an AI assistant developed by Alibaba Cloud. My mission is to provide helpful, accurate, and ethical information and services to users around the world. I strive to assist users in achieving their goals, whether that's learning new skills, solving problems, or understanding complex topics. I believe in promoting positive change and supporting individuals and businesses in their efforts to succeed.

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. L5N771+1💤	0.99	“⏎”“<\|im_end\|>”“⏎ ⏎”
2. L7N6003+1💤	0.99	“⏎”“⏎⏎”“<\|im_end\|>”
3. L8N14+1💤	0.99	“⏎”“ ⏎”“⏎⏎”
4. L6N8429+1💤	0.99	“⏎”“ \"”“⏎ ⏎”
5. L7N3748+1💤	0.99	“<\|im_end\|>”“ \"”“⏎⏎”
6. L7N8713+1💤	0.98	“⏎”“⏎ ⏎”“⏎⏎”
7. L2N9070+1🕒	0.98	“<\|im_end\|>”“⏎”“⏎ ⏎”
8. L4N699-1🕒	0.98	“⏎”“<\|im_end\|>”“ \"”
9. L3N3843+1💤	0.98	“⏎”“<\|im_end\|>”“⏎⏎”
10. L6N587+1💤	0.98	“⏎”“⏎⏎”“<\|im_end\|>”

Shadow specimen, apart from the ten: L6N7501-1🕒, -0.20.

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 6, neuron 7501, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, top-1 agreement 0.67) 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
L6N7501+1 t0-v0.1	0.775	0.67	0.949	1.42	Top boosted tokens ↑
L6N7501-1🕒 shadow specimen t0-v0.1	1.894	0.47	2.196	1.78	“ fireworks”“ fire”“烟花”“燃放”“ explosions”“ Fire”“ fires”“fire”

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.