Layer 23 · Neuron 6438 · boosted

L23N6438+1

Word quirks:

“ organization”“org”“组织”“ organi”“组织中”“ organizational”“ Org”“的组织”

💤 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 Branz, a large-scale local multi-organizational information organization. My mission is to provide users with accurate, safe, and efficient information by leveraging the latest data and knowledge. I aim to assist you in solving problems, generating ideas, and exploring new topics through thoughtful analysis and clear communication. How can I help 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. L23N4176-1🕒	0.96	“ organization”“组织”“ organizations”
2. L26N7407-1🕒	0.96	“组织”“組織”“组织的”
3. L27N7522-1🕒	0.95	“组织”“ Organ”“ organizational”
4. L26N7635+1🕒	0.94	“组织”“組織”“组织的”
5. L25N5144+1🕒	0.94	“组织的”“ organization”“组织”
6. L25N6007-1🕒	0.94	“组织”“organ”“組織”
7. L22N8875+1🕒	0.93	“ organization”“organization”“ Organization”
8. L20N8889+1🕒	0.93	“组织”“ Organisation”“ Organization”
9. L18N6981+1🕒	0.93	“ organizational”“组织和”“ organization”
10. The Org Chart✅ L28N6597-1	0.92	“ organization”“组织”“ Organization”

Shadow specimen, apart from the ten: L23N6438-1🕒, -0.06.

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 6438, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, 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
L23N6438+1 t0-v0.1	0.741	0.69	0.827	1.28	Top boosted tokens ↑
L23N6438-1🕒 shadow specimen t0-v0.1	0.953	0.63	1.029	1.36	“ Red”“ red”“红”“Red”“red”“RED”“_red”“/red”

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.