Layer 17 · Neuron 5964 · suppressed

L17N5964-1

Word quirks:

“ **“”“ **„”“ **‘”“ **"”“ **”“ **«”“ **'”“ **-”

🕒 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 Qwen3.1, the latest iteration in the Qwen series developed by Alibaba Cloud's Tongyi Lab. Here's a breakdown of my identity and capabilities:

Who Am I?

I am an advanced large language model designed to handle a wide spectrum of tasks. My foundation is built on a robust architecture that includes:

Contextual Understanding:
Enhanced Context Window: I can process and understand long-form content seamlessly.
Deep Reasoning: My training data spans up to 2024-09, allowing me

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. L10N7836-1💤	0.99	“ ”“ “”“ **"”
2. L17N7743+1🕒	0.99	“ ”“ '”“ **“”
3. L16N665+1💤	0.99	“ “”“ "”“ **”
4. L14N2004-1🕒	0.99	“ “”“ ‘”“ **”
5. L16N6484+1🕒	0.99	“ “”“ ”“ **‘”
6. L16N6904+1💤	0.99	“ “”“ ”“ **'”
7. L6N4877+1💤	0.99	“ “”“ ”“ **‘”
8. L15N6038-1💤	0.99	“ “”“ "”“ **”
9. L16N1853+1💤	0.99	“ "”“ “”“ **”
10. L9N2895+1💤	0.99	“ "”“ “”“ **”

Shadow specimen, apart from the ten: L17N5964+1🕒, 0.07.

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 5964, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, top-1 agreement 0.58) 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
L17N5964+1🕒 shadow specimen t0-v0.1	0.784	0.68	0.962	1.04	“.]”“ now”“.</”“.**”“ final”“ tutor”“ finally”“ thereafter”
L17N5964-1 t0-v0.1	0.756	0.58	0.912	1.36	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.