Layer 21 · Neuron 886 · suppressed

L21N886-1

Word quirks:

“Repair”“ minor”“ Equipment”“repair”“ equipment”“灾”“ repar”“新农”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Model 1, a general-purpose artificial intelligence assistant developed by General Model. My mission is to provide high-quality, safe, and helpful assistance across a wide range of tasks and domains.

Here is how I aim to fulfill this mission:

Helping with Tasks: Whether it's solving complex problems, generating creative content, analyzing data, or performing calculations, I strive to provide accurate and efficient solutions.
Enhancing Creativity: I can help you brainstorm ideas, write code, draft documents, or explore new concepts to spark innovation.
Promoting Safety & Ethics:

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. L30N1739+1💥	0.76	“修复”“ repair”“ Repair”
2. L24N2011+1💥	0.75	“ repair”“ repairs”“ Repair”
3. L23N2830+1🕒	0.75	“维修”“ repair”“ Repair”
4. L26N6128-1🕒	0.75	“ repairs”“ Repair”“ repair”
5. L26N647-1🕒	0.74	“ repair”“ repairs”“ Repair”
6. L23N8526-1🕒	0.74	“修复”“repair”“ Repair”
7. L24N6793+1🕒	0.74	“ repair”“repair”“修复”
8. L17N8033-1🕒	0.74	“修复”“ repairs”“ repair”
9. L19N8388-1🕒	0.73	“ repair”“维修”“修理”
10. L29N2555+1💤	0.73	“ repair”“ Reparatur”“修”

Shadow specimen, apart from the ten: L21N886+1🕒, 0.24.

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 886, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L21N886+1🕒 shadow specimen t0-v0.1	1.024	0.64	1.143	1.41	“ hollow”“ oily”“ thin”“ flattened”“-vertical”“ flat”“扁平”“悬”
L21N886-1 t0-v0.1	1.027	0.64	1.180	1.29	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.