Layer 28 · Neuron 4556 · boosted

L28N4556+1

Word quirks:

“ organization”“ organisation”“ Organization”“organization”“Organization”“ Organisation”“组织”“organisation”

💤 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 latest large language model developed by **T-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

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. L22N8875+1🕒	0.98	“ organization”“organization”“ Organization”
2. L23N1162-1💤	0.96	“ organization”“ Organization”“ free”
3. L20N8889+1🕒	0.96	“组织”“ Organisation”“ Organization”
4. The Org Chart✅ L28N6597-1	0.95	“ organization”“组织”“ Organization”
5. L20N743-1🕒	0.94	“ organizations”“ organisations”“ organization”
6. L23N4176-1🕒	0.94	“ organization”“组织”“ organizations”
7. L26N8537-1🕒	0.93	“ Organization”“Organization”“ Organisation”
8. L25N5144+1🕒	0.93	“组织的”“ organization”“组织”
9. L2N4601+1💥	0.93	“<\|im_end\|>”“ organization”“ organisation”
10. L0N5354-1💥	0.92	“ organisation”“<\|im_end\|>”“ tissue”

Shadow specimen, apart from the ten: L28N4556-1💤, 0.16.

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 28, neuron 4556, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.32, top-1 agreement 0.81) 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
L28N4556+1 t0-v0.1	0.321	0.81	0.359	1.23	Top boosted tokens ↑
L28N4556-1💤 shadow specimen t0-v0.1	0.313	0.82	0.325	1.30	“ Orig”“ orig”“虎”“扎”“ذ”“ORD”“原版”“ords”

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.