Layer 7 · Neuron 7142 · boosted

L7N7142+1

Word quirks:

“⏎”“ ____”“ ⏎”“ ⏎”“ ⏎⏎”“ ****”“ ——”“<|im_end|>”

🕒 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 in the AI Index Report, focused on general knowledge enhancement. My research is devoted to discovering new insights, publishing high-quality papers, and driving advancements in AI technology.

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. The Obliging Confabulator✅ L9N7468-1	0.94	“ ⏎”“ ⏎⏎”“<\|im_end\|>”
2. L7N3578+1💤	0.94	“<\|im_end\|>”“ ⏎⏎”“⏎⏎”
3. L11N4762-1💤	0.94	“<\|im_end\|>”“⏎”“ ⏎”
4. L11N1047+1⛔	0.94	“ ⏎⏎”“ ⏎”“⏎⏎”
5. L8N8380+1🕒	0.93	“<\|im_end\|>”“ ⏎”“ ——”
6. L7N3613-1🕒	0.93	“<\|im_end\|>”“⏎”“ ⏎”
7. L7N2250+1💤	0.93	“⏎”“ ⏎”“<\|im_end\|>”
8. L9N1397+1🕒	0.93	“ ⏎”“⏎”“ ⏎⏎”
9. L1N8788-1💥	0.92	“<\|im_end\|>”“ ⏎”“⏎”
10. L3N9174+1🕒	0.92	“ ⏎”“ ⏎⏎”“<\|im_end\|>”

Shadow specimen, apart from the ten: L7N7142-1🕒, 0.43.

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 7142, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.75, top-1 agreement 0.47) 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
L7N7142+1 t0-v0.1	1.750	0.47	2.079	2.13	Top boosted tokens ↑
L7N7142-1🕒 shadow specimen t0-v0.1	1.261	0.56	1.502	1.73	“...””“ -”“ --”“<\|im_end\|>”“ ‘”“ ””“ now”“!””

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.