Layer 23 · Neuron 9112 · suppressed

L23N9112-1

Word quirks:

“inky”“賓”“ Prec”“ Digital”“ПС”“람”“orno”“延展”

💤 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 Qianyi, a large language model developed by Qihin Digital. My primary mission is to provide high-quality, accurate, and helpful information and services to users like you. I aim to assist with answering questions, generating text, analyzing data, and supporting various tasks across different fields such as business, education, and creative work.

My design emphasizes:

Precision: Delivering factually accurate and contextually appropriate responses.
Responsiveness: Adapting to user needs while adhering to ethical guidelines and safety standards.
Inclusivity: Ensuring accessible and non-discrimative

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. L22N2813-1💤	0.67	“xis”“prop”“重复”
2. L31N1520-1💥	0.66	“reon”“拋”“PET”
3. L27N6210+1💤	0.66	“amel”“PK”“_mk”
4. L26N8738-1🕒	0.65	“ P”“紫”“P”
5. L29N8831+1🕒	0.65	“ipl”“平”“يب”
6. L25N2702+1💤	0.65	“棱”“онер”“พ”
7. L27N3766-1💤	0.64	“ Pena”“-p”“ipple”
8. L25N6984+1🕒	0.64	“emap”“Py”“ Sonic”
9. L27N8294-1💤	0.64	“oken”“дика”“yun”
10. L30N8459-1🕒	0.64	“ پاس”“okon”“зор”

Shadow specimen, apart from the ten: L23N9112+1🕒, 0.28.

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 23, neuron 9112, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, top-1 agreement 0.69) 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
L23N9112+1🕒 shadow specimen t0-v0.1	1.403	0.59	1.533	1.43	“ worship”“ Worship”“崇拜”“ worsh”“偶像”“ idol”“神”“ Idol”
L23N9112-1 t0-v0.1	0.708	0.69	0.761	1.30	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.