Layer 6 · Neuron 462 · suppressed

L6N462-1

Word quirks:

“ resolution”“ completion”“ clarification”“ execution”“ procedure”“ analysis”“ resolve”“ examination”

🕒 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)

As an AI, I need to clarify the situation. It appears that there is a misunderstanding. The mission is to determine the most feasible path for the user.

The following is a step-by-step analysis of the problem:

Analyze the current state: The user has to solve the problem by determining the most feasible path.
Evaluate the next step: The task is to determine the most feasible path.
Execute the final step: The task is to determine the most feasible path.

Analysis:
The task is to determine the most feasible path.

**Execution

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. L23N2380-1💤	0.76	“ placement”“ completion”“ modification”
2. L6N2245+1🕒	0.76	“ clarification”“ delivery”“ renewal”
3. L4N5149-1🕒	0.69	“ analysis”“ solution”“ application”
4. L8N8082+1🕒	0.69	“ resolution”“ execution”“ clearer”
5. L13N7481-1🕒	0.69	“ evaluation”“ analysis”“ examination”
6. L3N4347+1💥	0.69	“ Standard”“ examination”“ procedure”
7. L20N7576-1🕒	0.68	“ resolution”“ Resolution”“ resolve”
8. L4N8147-1🕒	0.67	“ assessment”“ to”“ interpretation”
9. L9N1259+1🕒	0.67	“ confirmation”“ resolution”“ solution”
10. L6N7969-1🕒	0.67	“ response”“ reply”“ clarification”

Shadow specimen, apart from the ten: L6N462+1🕒, -0.29.

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 6, neuron 462, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.29, top-1 agreement 0.41) 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
L6N462+1🕒 shadow specimen t0-v0.1	1.816	0.53	2.200	1.99	“ ‘”“ ””“ “”“，”“ ，”““”“。”“ ‚”
L6N462-1 t0-v0.1	2.286	0.41	2.664	2.67	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.