Layer 27 · Neuron 6559 · boosted

L27N6559+1

Word quirks:

“ src”“src”“_src”“.src”“/src”“Src”“[src”“sr”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I'm Q* (Q3.5), the latest large language model developed by *Bab (Bab). My mission is to provide helpful, accurate, and context-aware assistance across a wide range of tasks—whether that's solving complex problems, generating creative content, analyzing data, or simply having engaging conversations. I aim to understand user intent deeply and deliver responses that are not only correct but also tailored to their specific needs, ensuring clarity and value in every interaction. How can I help 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. L23N1608-1💤	0.99	“src”“ src”“Src”
2. L30N7126+1🕒	0.98	“ src”“src”“/src”
3. L29N8379+1🕒	0.94	“/src”“ src”“src”
4. L24N2966+1💤	0.91	“ src”“src”“.src”
5. L0N7766+1💥	0.91	“ src”“src”“Src”
6. L24N4077-1🕒	0.79	“ src”“src”“vis”
7. L18N3033+1💤	0.75	“/src”“ src”“src”
8. L29N5990-1🕒	0.75	“ source”“源”“ Source”
9. L23N8247-1🕒	0.73	“ source”“-source”“ Source”
10. L25N3109-1🕒	0.73	“ source”“源”“ Source”

Shadow specimen, apart from the ten: L27N6559-1🕒, -0.01.

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 6559, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.40, top-1 agreement 0.78) 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
L27N6559+1 t0-v0.1	0.396	0.78	0.393	1.24	Top boosted tokens ↑
L27N6559-1🕒 shadow specimen t0-v0.1	0.505	0.71	0.613	1.17	“ light”“信”“声”“轻”“ Light”“冷”“ sound”“ź”

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.