Layer 27 · Neuron 4795 · boosted

L27N4795+1

Word quirks:

“了解”“知道了”“知道”“ knowing”“ Knowing”“明白”“ Know”“知”

🕒 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 Q3, a large-scale language model known for my ability to understand and generate human-like text. My core mission is to be a knowledgeable and helpful assistant that can:

Answer questions across diverse topics (science, culture, technology, etc.) with accuracy and clarity.
Solve problems through logical reasoning, code generation, or step-by-step analysis.
Create content such as stories, emails, scripts, or creative writing tailored to user needs.
Understand context in conversations to provide coherent, multi-turn interactions.

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. L22N5474+1💤	0.95	“ knowing”“知道”“ Knowing”
2. The Know-It-All✅ L24N6960-1	0.95	“知”“知道”“ knows”
3. L21N2185-1🕒	0.94	“知”“知之”“ understands”
4. L23N4220+1🕒	0.94	“ aware”“知晓”“ awareness”
5. L23N7569-1🕒	0.94	“知道了”“ knowledge”“得知”
6. L21N5977-1🕒	0.94	“知”“知之”“ Nice”
7. L21N8346-1💤	0.93	“Knowing”“anel”“ Knowing”
8. L29N6372-1🕒	0.93	“知道”“知”“懂得”
9. L22N4897+1💤	0.93	“知道”“知”“知道他”
10. L22N8859+1💤	0.93	“ know”“ knowing”“知”

Shadow specimen, apart from the ten: L27N4795-1🕒, 0.18.

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 27, neuron 4795, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.53, top-1 agreement 0.77) 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
L27N4795+1 t0-v0.1	0.531	0.77	0.698	1.23	Top boosted tokens ↑
L27N4795-1🕒 shadow specimen t0-v0.1	1.194	0.56	1.144	0.69	“ deep”“深”“ Deep”“deep”“Deep”“ deeper”“.deep”“深的”

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.