Layer 23 · Neuron 1250 · boosted

L23N1250+1

Word quirks:

“ Hyp”“ hyp”“hyp”“控”“ hip”“控制”“-control”“ Control”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I'm Grok-2025, a large AI model trained by Eliteship. My mission is to help users like you explore ideas, learn new things, and create meaningful projects—whether it's coding, analyzing data, or just having a conversation. I'm designed to be flexible, creative, and always ready to adapt to your needs. 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. L26N4825-1🕒	0.74	“ control”“控制”“ Control”
2. L24N3427-1🕒	0.74	“ control”“ Control”“Control”
3. L24N8484-1💤	0.73	“控”“-control”“Control”
4. L25N7130+1🕒	0.73	“ control”“ Control”“控制”
5. L21N4885-1🕒	0.73	“ control”“控制”“ Control”
6. L26N5324-1🕒	0.73	“ control”“control”“ Control”
7. L26N5711-1🕒	0.73	“控制”“ control”“control”
8. L26N8417+1💥	0.73	“控制”“ control”“control”
9. L29N6487+1💤	0.73	“控制”“ control”“的控制”
10. L25N3499+1🕒	0.73	“ control”“控制”“control”

Shadow specimen, apart from the ten: L23N1250-1🕒, -0.28.

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 1250, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.67, top-1 agreement 0.71) 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
L23N1250+1 t0-v0.1	0.671	0.71	0.776	1.26	Top boosted tokens ↑
L23N1250-1🕒 shadow specimen t0-v0.1	0.811	0.68	0.849	1.29	“ (”“ («”“ ("”“ (`”“ ('”“ ((”“ $(”“ ([”

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.