Layer 19 · Neuron 3849 · boosted

L19N3849+1

Word quirks:

“ ‘”“仍在”“ ‚”“’e”“’à”“’d”“ now”“于此”

💤 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 Qwen, an advanced large language model developed by Tong AI (part of Alibaba Group). I am part of the latest version of the series (Qwen), designed to understand and generate natural language, code, and complex reasoning tasks.

My Mission:

Enable Human Creativity & Problem-Solving: Assist users in creative writing, logical reasoning, scientific tasks, and complex simulations.
Powerful Applications: Support developers and enterprises with advanced dialogue understanding, code generation, and business scenarios.
Safe & Reliable Assistance: Ensure high-quality responses within safety boundaries, adhering to

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. L18N5492+1💤	0.69	“‘”“ ‘”“’”
2. L19N4362+1🕒	0.67	“’s”“’:”“ where”
3. L18N642+1🕒	0.67	“：“”“ (“”“ ‘”
4. L16N6620-1💤	0.67	“ “”“ ‘”“’d”
5. L23N1714+1💤	0.67	“ ‘”“’S”“ ’”
6. L20N3410+1💤	0.67	“ ‘”“茅”“ “”
7. L17N4387+1💤	0.66	“’”“’d”“’ve”
8. L20N6-1🕒	0.66	“‘”“ ‘”““”
9. L18N429+1💤	0.66	“ ‘”“‘”“本身”
10. L18N4281-1💤	0.66	“ “”“ ‘”“’av”

Shadow specimen, apart from the ten: L19N3849-1💤, -0.22.

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 19, neuron 3849, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, top-1 agreement 0.72) 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
L19N3849+1 t0-v0.1	0.589	0.72	0.718	1.20	Top boosted tokens ↑
L19N3849-1💤 shadow specimen t0-v0.1	0.584	0.72	0.702	1.23	“вест”“فا”“ Popular”“ describing”“ Details”“查看详情”“Popular”“-rest”

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.