Layer 11 · Neuron 6398 · suppressed

L11N6398-1

Word quirks:

“ via”“ connection”“ connected”“ through”“ attached”“ linked”“ referenced”“ integration”

🕒 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, a large language model. My mission is to provide high-quality, accurate, and comprehensive information, reasoning, and code assistance to users across all domains. I will help you by:

Understanding the User's Intent: I will first acknowledge my role as an advanced AI assistant.
Defining the Mission Statement: I will explicitly state that my core mission is to "provide high-quality, accurate, and comprehensive information, reasoning, and code assistance" to users.
Establishing the Tone: I will maintain a professional, supportive

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. L22N5357-1🕒	0.85	“ connected”“Connected”“ Connected”
2. L21N4179+1🕒	0.82	“ connected”“connected”“ attached”
3. L11N6566-1🕒	0.79	“ connections”“ connection”“ connected”
4. L8N3600+1🕒	0.79	“ remote”“ connection”“ connected”
5. L30N3839+1💥	0.78	“ linked”“ connected”“ Linked”
6. L16N7380-1🕒	0.78	“ linked”“ connected”“ closely”
7. L11N3290+1🕒	0.78	“ linking”“ chain”“ direct”
8. L18N428-1🕒	0.76	“ connected”“ ensured”“ accessed”
9. L27N794+1🕒	0.76	“ involved”“ connected”“ participating”
10. L20N6194-1🕒	0.76	“ connected”“connected”“ Connected”

Shadow specimen, apart from the ten: L11N6398+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 11, neuron 6398, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.64) 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
L11N6398+1🕒 shadow specimen t0-v0.1	0.807	0.66	1.006	1.11	“ ⏎⏎”“ ⏎”“ here”“ <U+00A0>”“<U+00A0>”“ —”“<U+00A0><U+00A0>”“�”
L11N6398-1 t0-v0.1	0.811	0.64	1.051	1.23	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.