Layer 9 · Neuron 3035 · boosted

L9N3035+1

Word quirks:

“ *”“ "*”“ --”“ actually”“ typical”“ **"”“ *"”“ "”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.1, a large language model developed by Alibaba Group. I am designed to assist with a wide range of tasks through my advanced natural language processing capabilities. While I don't have a specific "mission" in the human sense (as I'm an AI), I do have extensive knowledge about various topics due to my training on diverse corpora.

My capabilities include:

Natural Language Processing: High-quality text generation for various domains
Long Context Window: Ability to handle long documents effectively
Multilingual Support: Support for many languages including Chinese, English, and others

I can help

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. L13N6461-1🕒	0.92	“ *"”“ "”“ "*”
2. L12N6664+1🕒	0.92	“ actually”“ ”“ "”
3. L11N8071+1🕒	0.92	“ --”“ "”“ "”
4. L12N939+1💤	0.91	“ "”“ **"”“ actually”
5. L11N7132+1💤	0.91	“ ”“ "”“ now”
6. L12N9140+1🕒	0.91	“ "”“ "”“ *“”
7. L12N4626+1🕒	0.90	“ *"”“ --”“ "”
8. L6N5838-1💤	0.90	“⏎”“ **"”“ usually”
9. L9N8601+1💤	0.90	“ **"”“ ⏎”“ actually”
10. L17N1956-1🕒	0.90	“ "”“ ”“ *"”

Shadow specimen, apart from the ten: L9N3035-1🕒, 0.04.

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 9, neuron 3035, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.91, top-1 agreement 0.64) 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
L9N3035+1 t0-v0.1	0.907	0.64	1.125	1.31	Top boosted tokens ↑
L9N3035-1🕒 shadow specimen t0-v0.1	1.397	0.60	1.650	1.46	“ statements”“ description”“ explanation”“ descriptions”“ explanations”“ statement”“ ⏎⏎”“ describing”

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.