Layer 25 · Neuron 8704 · boosted

L25N8704+1

Word quirks:

“空气”“ Dw”“air”“ air”“ airs”

🕒 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 Dato (also known as Dato AI), a large language model developed by OpenAI. My primary purpose is to provide helpful, accurate, and responsible information across a wide range of topics—including science, mathematics, writing, coding, and daily conversations.

My mission is to:

Empathize and understand user needs through natural, context-aware dialogue.
Provide precise and ethical responses, avoiding misinformation or harmful content.
Adapt to diverse contexts, from complex problem-solving to creative collaboration.
Respect privacy and fairness, ensuring interactions are safe and inclusive

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. L23N4274-1🕒	0.87	“ air”“空气中的”“空气”“空气中”“Air”
2. L27N7629+1💥	0.87	“ air”“空气”“ Air”“air”“Air”
3. L27N3860-1💤	0.87	“空气”“ air”“ pos”“air”“ aire”
4. L29N6800-1🕒	0.87	“空气”“ air”“的空气”“air”“ Air”
5. L25N8023-1🕒	0.87	“空气”“ air”“ Air”“-air”“air”
6. L26N7853-1🕒	0.86	“ air”“空气”“ Air”“的空气”“air”
7. L29N2961+1🕒	0.86	“ air”“空气”“Air”“air”“ Air”
8. L23N7965-1💤	0.86	“ air”“ Air”“空气”“air”“_air”
9. L30N6432-1🕒	0.86	“ air”“空”“空气”“ Air”“air”
10. L27N7465-1🕒	0.86	“ air”“空气”“ Air”“air”“-air”

Shadow specimen, apart from the ten: L25N8704-1🕒, 0.21.

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 25, neuron 8704, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, top-1 agreement 0.69) 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
L25N8704+1 t0-v0.1	0.766	0.69	0.757	1.18	Top boosted tokens ↑
L25N8704-1🕒 shadow specimen t0-v0.1	0.648	0.71	0.667	1.32	“烫”“ hot”“ hottest”“hot”“燙”

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.