Layer 21 · Neuron 451 · suppressed

L21N451-1

Word quirks:

“ first”“ last”“ original”“ same”“ usual”“ latter”“ origins”“usual”

🕒 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 Q3, a large language model developed by Qianxin (Qianxin Technology).

Who Am I?

I am designed to assist users in various tasks, including answering questions, generating text, coding, logical reasoning, and creative writing. I aim to provide clear, accurate, and context-aware responses while adhering to principles of safety and ethical guidelines.

Mission

My mission is to empower users by:

Providing reliable information and insights.
Supporting productivity through efficient problem-solving.
Encouraging creativity and innovation.
Ensuring interactions are safe

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. L21N37+1🧪	0.77	“ same”“same”“ oldest”
2. The Last Word✅ L20N837+1	0.68	“ source”“ last”“ origin”
3. L18N2600+1💤	0.68	“ original”“ latter”“ origin”
4. L17N5282-1💤	0.68	“ original”“-original”“ common”
5. L19N524-1🕒	0.67	“ same”“ latter”“ usual”
6. L18N496+1🕒	0.67	“ usual”“ difference”“ original”
7. L11N6016+1🕒	0.67	“ original”“ originally”“ originals”
8. L19N2882-1💤	0.66	“ consistently”“ originals”“ original”
9. L17N8013+1🕒	0.66	“ original”“ similar”“ same”
10. L15N8764-1🕒	0.66	“ now”“ basic”“ further”

Shadow specimen, apart from the ten: L21N451+1🕒, 0.30.

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 21, neuron 451, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.98, top-1 agreement 0.70) 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
L21N451+1🕒 shadow specimen t0-v0.1	2.433	0.41	2.609	1.59	“ some”“ none”“ those”“ part”“ one”“some”“ many”“ much”
L21N451-1 t0-v0.1	0.979	0.70	1.075	1.58	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.