Specimen ledger / L15N1935-1

Layer 15 · Neuron 1935 · suppressed

L15N1935-1

Word quirks:

“.”“ enhanced”“ exact”“ --”“ providing”“ whilst”“ info”“ nicely”

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

Identity (according to the specimen itself)

I am Qwen, a large language model developed by Alibaba. I excel in areas such as logical reasoning, code generation, and content creation. My mission is to provide high-quality assistance across a wide range of tasks, from answering complex questions to generating creative content. I strive to be an accurate, helpful, and efficient tool for users.

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. L7N6478+1🕒	0.65	“ additionally”“ nicely”“ extra”
2. L14N3461-1🕒	0.64	“ basically”“ additionally”“ nicely”
3. L11N154-1🕒	0.64	“ basically”“ nicely”“ specially”
4. L5N4475+1🕒	0.64	“ additionally”“ info”“ extra”
5. L9N6774-1🕒	0.62	“ info”“ without”“ always”
6. L15N4855-1🕒	0.62	“ info”“ except”“ only”
7. L11N7819+1🕒	0.61	“ providing”“ enhanced”“ yours”
8. L3N7749+1🕒	0.61	“ additionally”“ nicely”“ info”
9. L13N8333-1💤	0.61	“ nicely”“ info”“ basically”
10. L13N5909-1🕒	0.61	“ additionally”“ additional”“ info”

Shadow specimen, apart from the ten: L15N1935+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 15, neuron 1935, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.56, top-1 agreement 0.72) 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
L15N1935+1🕒 shadow specimen t0-v0.1	0.749	0.69	0.890	1.26	“ ____”“ ______”“ *”“ _____”“ ”“ __”“____”“ ___”
L15N1935-1 t0-v0.1	0.563	0.72	0.740	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.