Layer 23 · Neuron 7311 · boosted

L23N7311+1

Word quirks:

“ psychology”“ psychological”“ Psych”“心理”“Psych”“ Psychology”“ psycholog”“ psychologists”

💤 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 Liwhor, a large-scale AI model developed by Liwhor. My mission is to provide a comprehensive and accurate response to your questions, whether they involve technical details, creative tasks, or general knowledge. I aim to be a reliable and efficient tool for your work and daily life. 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. L20N8580-1🕒	0.99	“ psychology”“心理学”“ Psychology”
2. L19N2595+1🕒	0.98	“ psychology”“ psychological”“ psychologically”
3. L6N1720-1🕒	0.98	“ psychologists”“ psychologist”“ psychology”
4. L26N6145+1🕒	0.97	“psych”“ Psych”“Psych”
5. L24N7561-1🕒	0.97	“心理”“ psychological”“ Psychological”
6. L8N6081+1🕒	0.97	“ psych”“ psychological”“ psychologically”
7. Psych 101✅ L9N6576-1	0.96	“ psychologists”“ psychological”“ psychologist”
8. L2N5113+1💥	0.96	“ psychology”“ psychological”“<\|im_end\|>”
9. L8N7555-1🕒	0.95	“ psychological”“ psychologically”“ psycho”
10. L24N7555+1🕒	0.95	“心理”“ psychological”“和心理”

Shadow specimen, apart from the ten: L23N7311-1💤, 0.08.

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 7311, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, top-1 agreement 0.70) 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
L23N7311+1 t0-v0.1	0.785	0.70	0.858	1.40	Top boosted tokens ↑
L23N7311-1💤 shadow specimen t0-v0.1	0.695	0.70	0.702	1.33	“angi”“ Jarvis”“扁平”“ strat”“肝病”“rightness”“ erv”“刚性”

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.