Layer 18 · Neuron 8063 · suppressed

L18N8063-1

Word quirks:

“ necessary”“ specifically”“ purpose”“ intended”“ useful”“ structure”“ your”“ for”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen, a large language model designed to assist users in understanding and processing information. My primary goal is to provide accurate, relevant, and helpful responses to support your needs.

Key Aspects:

Purpose: To facilitate learning, problem-solving, and efficiency across diverse domains like technology, communication, and application development.
Functionality: I aim to balance accuracy with practicality, ensuring that interactions are meaningful and tailored to user intent.
Ethics: My design prioritizes ethical considerations, focusing on providing useful and positive outcomes for all users.

How would you like

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. L12N8211-1🕒	0.75	“ intent”“ appropriate”“ necessary”
2. L16N7367-1💤	0.73	“ intended”“ needed”“ meant”
3. L18N1880-1💤	0.70	“ necessary”“ pertinent”“ Necessary”
4. L13N8269+1🕒	0.70	“ purpose”“⏎”“ needed”
5. L11N7201-1🕒	0.70	“ intended”“ specifically”“ purpose”
6. L7N8664-1🕒	0.70	“ purpose”“ purposes”“ to”
7. L10N7290+1🕒	0.69	“ specific”“ defining”“ useful”
8. L7N4012-1💤	0.68	“ needed”“ purpose”“⏎”
9. L15N5915+1🕒	0.68	“ useful”“ if”“ specific”
10. L11N4474-1🕒	0.68	“ designed”“ always”“ purpose”

Shadow specimen, apart from the ten: L18N8063+1🕒, 0.10.

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 18, neuron 8063, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L18N8063+1🕒 shadow specimen t0-v0.1	0.828	0.62	1.023	1.73	“ represented”“ killer”“ spaces”“-space”“ space”“ depicted”“ bikes”“ health”
L18N8063-1 t0-v0.1	1.163	0.59	1.212	1.22	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.