Layer 29 · Neuron 4228 · boosted

L29N4228+1

Word quirks:

“ Quality”“quality”“和质量”“Quality”“质量”“-quality”“質”“_quality”

💤 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 Qwen3.5, the large-quality large language model developed by Tongyi Lab. My core mission is to be a capable, reliable, and safe AI assistant that empowers users across diverse scenarios. Here's how I strive to fulfill this:

Key Objectives:

Precision & Quality:

Deliver accurate, context-aware responses in text, code, data analysis, and creative tasks.
Support 100+ languages and handle complex queries with reduced hallucinations.

Efficiency & Speed:

Optimize for

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. L27N7013-1🕒	0.99	“ Quality”“Quality”“ quality”
2. Quality Control✅ L21N2219+1	0.98	“ quality”“ Quality”“Quality”
3. L25N2535+1🕒	0.98	“ quality”“quality”“Quality”
4. L27N8896-1🕒	0.98	“ quality”“ Quality”“Quality”
5. L28N5027+1🕒	0.97	“ quality”“ Quality”“Quality”
6. L19N6628-1🕒	0.97	“ quality”“ Quality”“Quality”
7. L25N5766-1🕒	0.97	“ quality”“ Quality”“质量”
8. L28N4542+1🕒	0.97	“质量”“的质量”“ quality”
9. L25N7318-1🕒	0.97	“ quality”“质量”“质量的”
10. L22N9100+1🕒	0.96	“ quality”“Quality”“quality”

Shadow specimen, apart from the ten: L29N4228-1🕒, 0.23.

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 29, neuron 4228, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.21, top-1 agreement 0.85) 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
L29N4228+1 t0-v0.1	0.213	0.85	0.242	1.28	Top boosted tokens ↑
L29N4228-1🕒 shadow specimen t0-v0.1	0.748	0.62	0.817	0.87	“高”“ high”“ highest”“-high”“highest”“最高”“high”“ Highest”

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.