Layer 14 · Neuron 2186 · suppressed

L14N2186-1

Word quirks:

“ now”“ then”“ here”“ these”“ elsewhere”“ also”“ further”“ NOW”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, a large language model developed by Alibaba Cloud. My mission is to empower users with knowledge, creativity, and efficiency across various domains. Here's how I strive to make a difference:

Versatile Assistance: From answering questions to generating code, I aim to streamline workflows.
Collaborative Support: I participate in competitions like COT-100K, helping teams refine their problem-solving skills.
Global Impact: By participating in events like the ACM ICPC, I contribute to technological advancements.
Real-world applications: For instance, I

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. L14N8334-1💤	0.88	“ now”“ further”“ then”
2. L14N6743-1🕒	0.83	“ also”“ here”“ additionally”
3. L14N7930-1🕒	0.82	“ also”“ here”“ this”
4. L14N8767-1💤	0.82	“ here”“ this”“ now”
5. L14N5208+1🕒	0.81	“ another”“ now”“ here”
6. L15N9120+1🕒	0.80	“ thus”“ now”“ here”
7. L15N953-1🕒	0.80	“ now”“ presently”“ additional”
8. L10N3448+1🕒	0.80	“ now”“ then”“<\|im_end\|>”
9. L14N4712+1🕒	0.80	“ presently”“ also”“ then”
10. L11N6923+1⛔	0.79	“ now”“ first”“ again”

Shadow specimen, apart from the ten: L14N2186+1🕒, 0.08.

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 14, neuron 2186, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.67) 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
L14N2186+1🕒 shadow specimen t0-v0.1	1.523	0.54	1.857	1.66	“ –”“ --”“ …”“……”“ —”“—in”“ […]”“?”
L14N2186-1 t0-v0.1	0.795	0.67	0.893	1.21	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.