Layer 11 · Neuron 7246 · boosted

L11N7246+1

Word quirks:

“ basically”“ essentially”“ actually”“ **"”“ similar”“ **”“ info”“ description”

🕒 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, a large-scale language model developed by Tongyi Lab. My primary mission is to assist users with a wide range of tasks, including:

Answering questions about facts, concepts, and general knowledge.
Solving math problems (e.g., arithmetic operations).
Generating text for creative writing, stories, or essays.
Assisting with logical reasoning and problem-solving.
Providing helpful summaries of information.
Supporting coding tasks (e.g., debugging code, optimizing performance).
**Handling

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 Paper Wall✅ L10N8384+1	0.79	“ actually”“ **"”“ clearer”
2. L10N8898+1🕒	0.78	“ \"”“ seems”“ info”
3. L11N4482-1🕒	0.77	“ "”“ basically”“ similar”
4. L10N2193+1💤	0.77	“ info”“ actually”“ description”
5. L12N2742+1💤	0.77	“ "”“ basically”“ description”
6. L14N6667-1💤	0.77	“ basically”“ “”“ "”
7. L9N5353-1🕒	0.76	“ basically”“ usually”“ **”
8. L12N4817-1💤	0.76	“ actually”“<\|im_end\|>”“ actual”
9. L12N4495+1🕒	0.75	“ basically”“ ____”“ really”
10. L13N1406-1🕒	0.75	“ here”“ basically”“ similar”

Shadow specimen, apart from the ten: L11N7246-1🕒, 0.11.

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 11, neuron 7246, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, 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
L11N7246+1 t0-v0.1	0.822	0.67	1.089	1.21	Top boosted tokens ↑
L11N7246-1🕒 shadow specimen t0-v0.1	0.917	0.66	1.093	1.39	“ ——”“ **“”“ ：”“ ，”“ ⏎”“ “”““”“ ‘”

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.