Layer 25 · Neuron 4826 · boosted

L25N4826+1

Word quirks:

“逻辑”“ logical”“ Logical”“logical”“ logically”“的逻辑”“Logical”“邏輯”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am GPT 01, a large language model trained by Zooa. My primary mission is to assist users in completing tasks through logical reasoning, creative content generation, and precise information processing. I aim to provide accurate, helpful, and contextually relevant responses while adhering to ethical guidelines and ensuring high-quality interactions across diverse applications. How can I assist 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. L23N4828-1🕒	0.98	“逻辑”“ Logic”“ logic”
2. L23N588+1🕒	0.97	“逻辑”“ logic”“ logically”
3. L20N5746-1🕒	0.97	“ logical”“逻辑”“ logic”
4. L17N8326-1🕒	0.97	“ logical”“ logically”“ logic”
5. L25N1778+1💤	0.97	“逻辑”“廉”“ logically”
6. L21N6529-1💥	0.96	“ logical”“逻辑”“ Logical”
7. L31N5062-1💥	0.96	“ logical”“ Logical”“Logical”
8. L31N2140-1💥	0.96	“逻辑”“ logic”“的逻辑”
9. L23N2439-1💤	0.96	“逻辑”“ Logic”“ Logical”
10. L23N3141-1🕒	0.95	“ logical”“逻辑”“ logically”

Shadow specimen, apart from the ten: L25N4826-1🕒, 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 25, neuron 4826, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, 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
L25N4826+1 t0-v0.1	0.679	0.69	0.688	1.42	Top boosted tokens ↑
L25N4826-1🕒 shadow specimen t0-v0.1	0.940	0.64	0.975	1.11	“^”“^{”“^{-”“ power”“^(”“幂”“ ^”“ powers”

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.