Layer 17 · Neuron 3004 · suppressed

L17N3004-1

Word quirks:

“ basically”“ again”“ suggests”“ probably”“ essentially”“ suggested”“ suggest”“ suggesting”

🕒 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 developed by Tongyi Lab. My name is Qwen3.5, and I can be considered an advanced AI assistant with the following core capabilities:

Advanced Language Understanding

I have been trained on a diverse dataset to handle complex tasks, including logical reasoning and multi-language support (over 100 languages). This allows me to understand and process text effectively.

Task Execution Capabilities

I can perform complex tasks such as planning and executing multi-step operations. For example, I can break down a task

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. L17N6592-1🕒	0.87	“ suggesting”“ suggest”“ possibly”
2. L13N1005+1🕒	0.86	“ possibly”“ suggested”“ suggesting”
3. L17N223-1💤	0.86	“ perhaps”“ presumably”“ suggests”
4. L13N794-1🕒	0.85	“ possibly”“ suggests”“ implying”
5. L15N6066-1💤	0.84	“ suggested”“ suggesting”“ therefore”
6. L20N1449-1🕒	0.84	“ likely”“ suggesting”“ usually”
7. L17N3307+1🕒	0.84	“ suggested”“ suggests”“ does”
8. L15N4483-1💤	0.83	“ similar”“ suggested”“ suggesting”
9. L16N2755-1🕒	0.83	“ suggestions”“ discussion”“ suggests”
10. L14N1685-1💤	0.83	“ usually”“ suggests”“ implies”

Shadow specimen, apart from the ten: L17N3004+1💤, 0.05.

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 3004, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.69) 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
L17N3004+1💤 shadow specimen t0-v0.1	0.542	0.67	0.581	1.14	“...\”“...*”“ ⏎⏎”“ [...]”“⏎⏎”“...</”“...)”“...””
L17N3004-1 t0-v0.1	0.698	0.69	0.893	1.14	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.