Layer 11 · Neuron 1568 · boosted

L11N1568+1

Word quirks:

“ currently”“ presently”“ recently”“ current”“ now”“ recent”“ lately”“ today”

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

Identity (according to the specimen itself)

I am an AI assistant designed to provide helpful, accurate information. My mission is to assist users with tasks that require reasoning, including complex problem-solving. I operate under a "three-step" framework:

Direct Answer: For simple questions, I give a direct response.
Principle: For questions requiring some thought but not deep calculation, I explain the underlying principle.
Process: For complex problems needing multi-step reasoning, I outline the process.
Step-by-Step: For problems requiring full calculation before reaching a final conclusion, I show the complete step

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. L15N3637-1🕒	0.98	“ currently”“ presently”“ today”
2. L12N3949-1🕒	0.98	“ currently”“ recently”“ now”
3. L12N4932+1🕒	0.95	“ current”“ recent”“ now”
4. L14N6563+1💤	0.95	“ now”“ currently”“ updated”
5. L15N7394+1💤	0.94	“ now”“ presently”“ currently”
6. L10N8526-1🕒	0.94	“ recently”“ now”“ similar”
7. L13N8554+1🕒	0.93	“ currently”“ current”“ presently”
8. L13N1334-1💤	0.93	“ now”“<\|im_end\|>”“ nowadays”
9. L11N8255+1💤	0.92	“ current”“ currently”“ specific”
10. L9N3747-1💤	0.92	“ currently”“ current”“ now”

Shadow specimen, apart from the ten: L11N1568-1🕒, 0.17.

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 11, neuron 1568, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, top-1 agreement 0.72) 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
L11N1568+1 t0-v0.1	0.744	0.72	0.899	1.39	Top boosted tokens ↑
L11N1568-1🕒 shadow specimen t0-v0.1	2.062	0.41	2.370	1.92	“ each”“ maybe”“ individually”“ ⏎”“ individual”“ EACH”“ ____”“ every”

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.