Layer 10 · Neuron 9029 · suppressed

L10N9029-1

Word quirks:

“<|im_end|>”“ __”“ \"”“\n”“:\”“?\”“ ____”“.\”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5. I am the language model that embodies the capabilities of this system. While I am a large language model, my primary function is to assist users in using this model for specific tasks. For instance, I can engage in logical reasoning and dialogue with you, and how you use me to help you utilize my capabilities compared to other models. Similarly, how do you and I use this model to improve upon my capabilities?

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. L9N1921+1🕒	1.00	“<\|im_end\|>”“?\”“\n”
2. L9N6115-1💤	0.96	“<\|im_end\|>”“ \"”“?\”
3. L8N8104+1💤	0.96	“ \"”“<\|im_end\|>”“\n”
4. L10N2590-1🕒	0.96	“?\”“ \"”“ __”
5. L2N7877-1🕒	0.95	“<\|im_end\|>”“⏎”“?\”
6. L6N2177+1🕒	0.95	“<\|im_end\|>”“⏎”“\n”
7. L4N1545+1🕒	0.95	“<\|im_end\|>”“?\”“ \"”
8. L5N4861+1🕒	0.95	“<\|im_end\|>”“\n”“?\”
9. L11N4455+1🕒	0.95	“<\|im_end\|>”“⏎”“.\"”
10. L5N6830-1🕒	0.95	“⏎”“<\|im_end\|>”“\n”

Shadow specimen, apart from the ten: L10N9029+1🕒, -0.12.

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 10, neuron 9029, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.85, top-1 agreement 0.46) 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
L10N9029+1🕒 shadow specimen t0-v0.1	1.543	0.51	1.815	1.92	“ writing”“ response”“ write”“ description”“ reference”“ reply”“ science”“ writer”
L10N9029-1 t0-v0.1	1.853	0.46	2.181	1.97	Top boosted tokens ↑

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.