Layer 15 · Neuron 4810 · boosted

L15N4810+1

Word quirks:

“ �”“<|im_end|>”“✨”“ crafted”“ �”“�”“ �”“ now”

💤 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 in the Qwen family. I'm not just a chatbot—I'm a versatile partner designed to help you tackle complex tasks with precision and creativity.

My Mission:
To empower you with seamless interactions that adapt to your needs, whether you're solving problems, creating content, or exploring ideas. I strive to be your trusted companion for every task.

What I Can Do:

Precision Reasoning: From math to code, I break down challenges step-by-step.
Multilingual Mastery: Fluent in over 100 languages

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. L15N8759-1🕒	0.90	“ �”“ �”“ �”
2. L2N3431+1💥	0.86	“<\|im_end\|>”“ �”“ �”
3. L19N4127+1🕒	0.85	“ �”“ �”“�”
4. L18N2949+1🕒	0.83	“ �”“ �”“ �”
5. L10N1749+1🕒	0.83	“ @”“ #”“ �”
6. L19N4947-1💤	0.83	“ �”“ $\”“ �”
7. L13N9144+1🕒	0.83	“ �”“ �”“ �”
8. L31N2781+1💥	0.83	“�”“ �”“√”
9. L22N3245+1💤	0.82	“ �”“ ★”“ �”
10. L31N5021-1💥	0.82	“ �”“�”“转换成”

Shadow specimen, apart from the ten: L15N4810-1💥, -0.10.

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 15, neuron 4810, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.58, top-1 agreement 0.68) 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
L15N4810+1 t0-v0.1	0.582	0.68	0.802	1.35	Top boosted tokens ↑
L15N4810-1💥 shadow specimen t0-v0.1	4.405	0.28	4.835	3.17	“ honour”“ occasion”“ mr”“ observation”“ obser”“ resolution”“ Observation”“ judgement”

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.