Layer 21 · Neuron 6028 · suppressed

L21N6028-1

Word quirks:

“ clearly”“ clearer”“清晰”“ clear”“ชัดเจน”“ Clearly”“明确”“ jelas”

🕒 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 GigaGPT, a large language model developed to provide clear, concise, and actionable information. My mission is to help you quickly find what you need—whether it's answering a specific question, clarifying complex topics, or providing practical guidance—without unnecessary detail. I aim to be direct and useful, ensuring you get straight to the point while respecting your time. How can I assist you today?

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. L22N1230-1🕒	0.94	“ clearly”“ clearer”“ clear”
2. L23N5757+1🕒	0.92	“ clearly”“ Clearly”“清晰”
3. L18N9013-1🕒	0.92	“ clearly”“ clear”“ clearer”
4. L21N9133+1🕒	0.91	“ clearly”“很明显”“ clear”
5. L17N7331-1🕒	0.91	“ clear”“ clearly”“清晰”
6. L18N7709-1🕒	0.90	“ clear”“ clearly”“清晰”
7. L21N3416+1💤	0.89	“ clear”“ obvious”“ rõ”
8. L20N139-1🕒	0.89	“ clear”“清晰度”“清晰”
9. L22N980+1🕒	0.88	“明确”“ Clearly”“ Clear”
10. L18N4672-1💤	0.88	“ clearly”“ clear”“ Clearly”

Shadow specimen, apart from the ten: L21N6028+1🕒, 0.04.

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 6028, 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 0.70) 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
L21N6028+1🕒 shadow specimen t0-v0.1	1.529	0.53	1.687	1.45	“问我”“ ask”“问”“ asking”“問”“问她”“问他”“ Ask”
L21N6028-1 t0-v0.1	0.800	0.70	0.919	1.27	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.