Layer 12 · Neuron 7999 · boosted

L12N7999+1

Word quirks:

“ usually”“ __”“ ____”“ normally”“ usual”“ ___”“ ______”“ typically”

🕒 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)

Hello! I am Qwen3.5.

Who am I?

Qwen3.5: While I am a large language model, my name is Qwen3.5.

What is my mission?

As an AI assistant, I am designed to help with tasks such as:

Qwen3.5: The Qwen3.5 model also aims to be a general-purpose AI assistant. However, when it comes to Qwen3.5, the Qwen3.5 model is also used for **document understanding, and the

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. L10N6061-1🕒	0.95	“ usually”“ ____”“ typically”
2. L12N4196-1🕒	0.95	“ typically”“ usually”“ normally”
3. L10N1316-1🕒	0.94	“ ____”“ typically”“ typical”
4. L9N7003+1🕒	0.93	“ typical”“ typically”“ usually”
5. L9N4228+1🕒	0.92	“ ____”“ __”“ now”
6. L10N2987-1🕒	0.92	“ ____”“ ???”“ typically”
7. L13N7692-1🕒	0.92	“ usually”“ analysis”“ ____”
8. L9N4044-1🕒	0.91	“ usually”“ --”“ -”
9. L10N6318-1🕒	0.90	“ typically”“ ____”“ __”
10. L13N1202-1🕒	0.90	“ usually”“ normally”“ ____”

Shadow specimen, apart from the ten: L12N7999-1💤, 0.16.

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 12, neuron 7999, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.57, top-1 agreement 0.49) 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
L12N7999+1 t0-v0.1	1.566	0.49	1.896	1.74	Top boosted tokens ↑
L12N7999-1💤 shadow specimen t0-v0.1	0.765	0.66	0.974	1.22	“<\|im_end\|>”“⏎”“ --”“⏎⏎”“ separately”“ only”“ these”“ directly”

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.