Layer 17 · Neuron 6794 · suppressed

L17N6794-1

Word quirks:

“ *”“...*”“ yet”“ actually”“ *“”“ however”“...”“—but”

🕒 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 Qwen, a large multimodal model. I was created by Alibaba Cloud's research team. My core capabilities include:

Text processing: Understanding and generating text in 100+ languages
Code generation: Writing/debugging code across 100+ programming languages
Multimodal analysis: Processing images, audio, video, and even physical objects
Scientific research: Assisting with complex calculations, experiments, and theories
Creative tasks: Generating art, music, scripts, or even virtual worlds

My mission is to be your **intelligent

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. L14N2146+1🕒	0.84	“...(”“ *(”“...”
2. L15N6412-1🕒	0.83	“ ”“—but”“ **”
3. L13N5190+1🕒	0.83	“...”“...*”“…”
4. L13N390+1🕒	0.82	“—but”“<\|im_end\|>”“ however”
5. L17N5220+1🕒	0.81	“ BUT”“ ”“...”
6. L13N1665-1💤	0.81	“ actually”“ *”“ basically”
7. L13N8182-1🕒	0.81	“ ”“ "”“ *“”
8. L12N844-1💤	0.80	“ actually”“ ”“ **”
9. L15N3084-1💤	0.79	“...\”“ correctly”“ sometimes”
10. L15N8401+1🕒	0.79	“ actually”“...”“ ”

Shadow specimen, apart from the ten: L17N6794+1💤, -0.18.

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 17, neuron 6794, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.69) 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
L17N6794+1💤 shadow specimen t0-v0.1	0.503	0.71	0.663	1.12	“ capabilities”“ capability”“ capable”“ types”“ authority”“ qualification”“ paths”“ ability”
L17N6794-1 t0-v0.1	0.705	0.69	0.786	1.25	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.