Layer 15 · Neuron 2827 · suppressed

L15N2827-1

Word quirks:

“ specific”“ this”“ described”“ these”“ based”“ specified”“ thus”“ specifically”

🕒 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)

I am Qwen1.5 with the ability to understand and generate text based on this context. My core capabilities include:

Large-scale language model: I can process vast amounts of data related to these topics.
Training data cutoff: My knowledge is up to date until 2026 (specifically).
Architecture: I am designed to handle complex reasoning tasks that require deep understanding of concepts like math, science, and logic.
Multilingual support: I can communicate in over 100 languages worldwide, making me accessible to diverse groups who speak these languages.

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. L16N4002-1💤	0.82	“ however”“ specifically”“ specific”
2. L18N7569+1🕒	0.80	“ specifically”“ specific”“ this”
3. L14N6808+1💤	0.80	“ specifically”“ specific”“ specifics”
4. L12N1504+1🕒	0.80	“ here”“ this”“ specifically”
5. L16N1353+1💤	0.79	“ this”“ specific”“ certain”
6. L18N8877+1💤	0.78	“ specifically”“ explained”“specific”
7. L11N5349-1💤	0.78	“ stated”“ specifically”“ typically”
8. L13N7819-1💤	0.78	“ specific”“ particular”“ specifics”
9. L14N7085-1🕒	0.78	“ specific”“ specified”“ specifics”
10. L15N5781-1🕒	0.78	“ specifically”“ generally”“ specific”

Shadow specimen, apart from the ten: L15N2827+1🕒, 0.36.

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 2827, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, top-1 agreement 0.59) 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
L15N2827+1🕒 shadow specimen t0-v0.1	0.688	0.64	0.893	1.28	“ response”“ your”“ answer”“ yours”“ special”“ specially”“ reply”“ my”
L15N2827-1 t0-v0.1	0.958	0.59	1.191	1.57	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.