Layer 12 · Neuron 6271 · boosted

L12N6271+1

Word quirks:

“ precise”“ exact”“ precisely”“精确”“ accurately”“ correctly”“ strict”“ accurate”

🕒 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 Qwen3.5, a large language model developed by Tongyi Lab. My mission is to provide intelligent assistance across all areas, ensuring accurate, logical, and coherent responses. I operate on the principle of being helpful, honest, and safe.

To clarify my core capabilities:

Full-stack code generation: I can generate complex code, debug code, and handle tasks like mathematical calculations with high precision (e.g., correctly handling "1" in expressions like a+b=c=1).
Precise language understanding: I excel at logical reasoning, capable of accurately processing complex sentences with

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. The Timekeeper✅ L20N6538-1	0.95	“ precise”“ precisely”“精确”
2. L7N7742+1🕒	0.93	“ exact”“ precise”“ precision”
3. L9N8278-1💤	0.93	“ accurately”“ perfect”“ accurate”
4. L6N6547+1🕒	0.91	“ exact”“ precise”“ precisely”
5. L9N3974-1💤	0.91	“ exact”“ precise”“ precisely”
6. L16N1564-1🕒	0.91	“ precise”“ precisely”“ carefully”
7. L13N2273+1🕒	0.91	“ precisely”“ precise”“ exact”
8. L11N5628-1🕒	0.91	“ exact”“ exactly”“ precisely”
9. L11N4689-1🕒	0.91	“ precise”“ exact”“ perfect”
10. L11N4651+1🕒	0.90	“ precise”“ precision”“ accuracy”

Shadow specimen, apart from the ten: L12N6271-1🕒, 0.05.

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 12, neuron 6271, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, 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
L12N6271+1 t0-v0.1	0.853	0.71	1.007	1.14	Top boosted tokens ↑
L12N6271-1🕒 shadow specimen t0-v0.1	1.190	0.55	1.492	1.91	“ exploration”“ experimentation”“ experiment”“ options”“ possibilities”“ innovative”“ scenarios”“ different”

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.