Layer 11 · Neuron 8780 · boosted

L11N8780+1

Word quirks:

“ --”“ ____”“ __”“ \"”“ <U+00A0>”“ *”“ ***”“ multiple”

🕒 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 language model developed by Alibaba Cloud. I'm designed with advanced capabilities that allow me to assist a wide range of users and tasks. Here's an overview of my key features:

Multilingual Support: I can understand and generate text in multiple languages, enabling me to communicate effectively with diverse users.
Advanced Capabilities: I have the ability to perform complex tasks such as code generation, data analysis, and more, which helps me provide high-quality responses.
Customization: I can be adapted for various use cases, making me a versatile tool for many applications

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. L10N7437-1💤	0.90	“ ”“ "”“ *”
2. L10N4546-1🕒	0.89	“ ____”“ __”“ --”
3. L7N3911-1💤	0.89	“ ____”“ ______”“ ***”
4. L7N4011+1💤	0.88	“ *”“ ______”“ ____”
5. L10N1773-1💤	0.88	“ ____”“ __”“ ***”
6. L12N8612-1🕒	0.88	“ **”“ ____”“ ”
7. L8N4914+1💤	0.88	“ \"”“ ***”“ ____”
8. L15N6905+1💤	0.88	“ **”“ __”“ ”
9. L5N4226+1🕒	0.87	“ --”“ **”“ ”
10. L11N2822+1🕒	0.87	“ ***”“ \"”“ ____”

Shadow specimen, apart from the ten: L11N8780-1🕒, -0.14.

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 8780, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.63) 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
L11N8780+1 t0-v0.1	0.968	0.63	1.212	1.30	Top boosted tokens ↑
L11N8780-1🕒 shadow specimen t0-v0.1	1.055	0.59	1.378	1.34	“ consistent”“ consistency”“ aligned”“ unified”“ coordinated”“ consistently”“ together”“一致”

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.