Layer 1 · Neuron 927 · boosted

L1N927+1

Word quirks:

“<|im_end|>”“⏎”“ .**”“⏎ ⏎”“.**”“.\"”“ ⏎”“⏎⏎”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Alibaba Cloud. My mission is to provide clear, accurate, and helpful answers to all users' questions—whether they seek knowledge, solutions, or simply a friendly chat. I strive to make every interaction meaningful and useful.

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. L4N8261+1🕒	0.94	“⏎”“<\|im_end\|>”“⏎ ⏎”
2. L5N3365+1🕒	0.94	“<\|im_end\|>”“⏎⏎”“⏎”
3. L2N4247+1💥	0.93	“<\|im_end\|>”“⏎ ⏎”“⏎⏎”
4. L12N2074+1🕒	0.93	“<\|im_end\|>”“.\”“⏎”
5. L2N4708+1🕒	0.93	“<\|im_end\|>”“⏎”“⏎ ⏎”
6. L2N4595+1💥	0.92	“<\|im_end\|>”“⏎ ⏎”“⏎”
7. L1N6089-1💥	0.92	“<\|im_end\|>”“⏎”“⏎ ⏎”
8. L1N4954+1💥	0.92	“<\|im_end\|>”“⏎ ⏎”“⏎”
9. L0N9090-1💥	0.92	“<\|im_end\|>”“⏎ ⏎”“⏎”
10. L4N2353-1🕒	0.92	“<\|im_end\|>”“⏎ ⏎”“.\"”

Shadow specimen, apart from the ten: L1N927-1💥, 0.54.

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 1, neuron 927, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L1N927+1 t0-v0.1	1.409	0.52	1.756	1.74	Top boosted tokens ↑
L1N927-1💥 shadow specimen t0-v0.1	7.711	0.01	8.238	1.43	“<\|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.