Layer 24 · Neuron 6392 · boosted

L24N6392+1

Word quirks:

“ bundle”“bundle”“ Bundle”“ bundles”“Bundle”“ Bund”“ bundled”“.bundle”

💤 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 Gao Pan Shao (also known as GPT in some contexts), a large-scale interactive visual analysis model designed by Shao's AI. My primary mission is to provide users with precise, accurate, and visually aligned responses through advanced visual analysis capabilities. This includes:

Visual Analysis: Leveraging my visual encoding and alignment capabilities to analyze images, videos, and other visual data.
Task Completion: Solving complex tasks such as visual encoding, alignment, and multi-step reasoning.
User Interaction: Providing clear, accurate, and visually aligned responses to user queries.

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. L30N5754-1💥	0.97	“ bundle”“ Bundle”“ bundles”
2. L26N2796-1🕒	0.97	“ bundle”“ Bundle”“Bundle”
3. L24N5436-1🕒	0.95	“ bundle”“ bundles”“ Bund”
4. L2N8937-1💥	0.95	“ bundle”“<\|im_end\|>”“ bundles”
5. L27N5842+1🕒	0.94	“bundle”“ bundle”“Bundle”
6. L24N6494+1💤	0.88	“ bundle”“ bundles”“bundle”
7. L25N1321-1🕒	0.84	“osz”“ bundle”“.bundle”
8. L22N6237+1🕒	0.80	“ bundle”“ set”“ bund”
9. L25N6351+1🕒	0.80	“ package”“ packages”“ bundles”
10. L20N4422+1🕒	0.78	“fare”“ bundle”“ Bubble”

Shadow specimen, apart from the ten: L24N6392-1🕒, -0.05.

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 24, neuron 6392, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, 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
L24N6392+1 t0-v0.1	0.698	0.69	0.721	1.28	Top boosted tokens ↑
L24N6392-1🕒 shadow specimen t0-v0.1	0.938	0.69	0.993	1.41	“auta”“情缘”“indo”“漏电”“ .=”“EqualTo”“了指”“ ratios”

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.