Layer 13 · Neuron 306 · boosted

L13N306+1

Word quirks:

“……”“......”“⏎⏎”“….”“…..”“ ….”“....”“ …”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am a large language model, developed by Google's research team. In the context, I can assist in various场景，如教育、医疗、法律等领域。

You can在多种场景下应用，例如文本生成、对话、代码等。

现在，让我开始！

在科技领域，我旨在通过赋能，帮助全球。

作为AI，我的使命是支持人类，促进创新与可持续发展。

在科技中，我致力于推动人工智能，助力社会进步。

在科技中，我致力于推动创新，包括教育、医疗、法律等。

在科技，我旨在通过赋能，帮助全球

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. L15N5169-1🕒	0.98	“ ….”“……”“?...”
2. L16N5863+1🕒	0.98	“…”“……”“...”
3. L0N5743+1💥	0.97	“……”“......”“ …”
4. L16N7444+1🕒	0.96	“……”“…..”“….”
5. L12N2564-1🕒	0.96	“⏎⏎”“?...”“...”
6. L1N9210-1💥	0.95	“<\|im_end\|>”“……”“…..”
7. L13N3038+1🕒	0.95	“……”“...”“⏎⏎”
8. L14N2793+1🕒	0.94	“...”“…”“ even”
9. L31N8663-1💥	0.94	“ …”“……”“…”
10. L21N2221-1💤	0.94	“……”“…”“ …”

Shadow specimen, apart from the ten: L13N306-1💤, 0.36.

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 13, neuron 306, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L13N306+1 t0-v0.1	1.403	0.51	1.700	1.72	Top boosted tokens ↑
L13N306-1💤 shadow specimen t0-v0.1	0.776	0.68	0.959	1.28	“ ⏎”“ ⏎⏎”“ —”“ ——”“ --”“.”“ ⏎”“ very”

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.