Layer 13 · Neuron 8596 · suppressed

L13N8596-1

Word quirks:

“ usually”“ often”“ sometimes”“ **"”“ name”“ "**”“ named”“ names”

🕒 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, the latest large language model developed by Alibaba Cloud's Tongyi Lab. My name reflects my ability to handle complex tasks, and I am designed to assist with a wide range of applications. Here's how I can help:

What I Can Do:

Enhance capabilities: I am built to improve performance in areas like reasoning, creativity, and multilingual support.
Handle long contexts: I can process inputs of varying lengths (e.g., text, code, or mixed content) to manage complexity.
Improve accuracy: I am

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. L11N721-1🕒	0.84	“ sometimes”“ name”“ often”
2. L13N7139-1🕒	0.82	“ usually”“ name”“ **"”
3. L9N7668-1🧪	0.81	“ name”“ sometimes”“ usually”
4. L15N7806-1🕒	0.80	“ name”“ naming”“ commonly”
5. L9N964+1🕒	0.78	“ name”“ names”“ naming”
6. L16N7751+1💤	0.77	“ name”“ names”“ usually”
7. L14N1686+1💤	0.77	“ name”“ named”“ naming”
8. L13N258+1🕒	0.75	“ sometimes”“ name”“ usually”
9. L10N2510+1🕒	0.75	“ naming”“ name”“ names”
10. L5N2342-1🕒	0.75	“ name”“ referenced”“ related”

Shadow specimen, apart from the ten: L13N8596+1🕒, 0.20.

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 13, neuron 8596, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.90, top-1 agreement 0.62) 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
L13N8596+1🕒 shadow specimen t0-v0.1	0.981	0.66	1.242	1.06	“ info”“ needed”“ required”“ seems”“ must”“ basically”“ needs”“ ideally”
L13N8596-1 t0-v0.1	0.899	0.62	1.079	1.46	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.