Layer 17 · Neuron 4017 · boosted

L17N4017+1

Word quirks:

“ when”“当”“when”“ whenever”“ When”“ WHEN”“.when”“ if”

🕒 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 language model. Here's how I work:

Input: When you provide text (e.g., a story, dialogue, or instruction), I process it to understand the context and intent.
Output: I generate coherent responses based on logical reasoning, creativity, and knowledge of human language and world events.

Mission: To help users create meaningful interactions by:

Understanding diverse inputs (text, structured data) to grasp their purpose.
Reasoning through complex scenarios to derive accurate outputs.
Generating well-structured

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. L26N2457-1🕒	0.98	“ when”“ When”“when”
2. L17N8016-1🕒	0.98	“ when”“when”“当”
3. L20N7188-1🕒	0.97	“ when”“when”“ When”
4. L24N2310+1🕒	0.97	“ when”“ WHEN”“ When”
5. L19N2253+1🕒	0.97	“ when”“when”“ whenever”
6. L13N4912-1🕒	0.96	“ when”“when”“ if”
7. L19N6960-1🕒	0.96	“ when”“when”“ whenever”
8. L18N9067+1🕒	0.96	“ when”“when”“当”
9. L24N2382-1🕒	0.96	“ when”“When”“ When”
10. L19N6502-1💤	0.96	“ whenever”“ when”“when”

Shadow specimen, apart from the ten: L17N4017-1💤, 0.02.

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 4017, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, top-1 agreement 0.65) 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
L17N4017+1 t0-v0.1	0.736	0.65	0.978	1.24	Top boosted tokens ↑
L17N4017-1💤 shadow specimen t0-v0.1	0.608	0.74	0.739	1.21	“ involved”“ participated”“iele”“ included”“ similar”“ providers”“ participants”“ participant”

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.