Layer 8 · Neuron 5868 · boosted

L8N5868+1

Word quirks:

“ instantly”“ now”“ quickly”“⏎”“ usually”“ immediately”“ quick”“ immediate”

💤 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 a large language model trained by Google. My mission is to provide helpful, accurate, and harmless information while adhering to ethical guidelines. I aim to assist users in solving problems, answering questions, and facilitating learning across a wide range of topics. Whether you need help with writing, research, or just general knowledge, I strive to deliver clear and concise responses that meet your needs.

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. L11N6295-1🕒	0.93	“ quick”“ now”“ immediate”
2. L15N4356-1🕒	0.93	“ immediate”“ immediately”“ quickly”
3. L9N6727+1🕒	0.91	“ instantly”“ immediate”“ quickly”
4. L7N2719+1💤	0.89	“ now”“ instant”“ instantly”
5. L15N6220+1🕒	0.89	“ instantly”“ immediately”“ immediate”
6. L17N8356+1🕒	0.88	“ instantly”“ immediate”“ immediately”
7. L11N3909-1🕒	0.88	“ immediate”“ immediately”“ initial”
8. L9N4606-1💤	0.88	“ immediately”“ instantly”“ immediate”
9. L9N7831-1🕒	0.88	“ instantly”“ instantaneous”“ instant”
10. L23N285-1🕒	0.88	“ instant”“ immediately”“instant”

Shadow specimen, apart from the ten: L8N5868-1🕒, 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 8, neuron 5868, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.67) 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
L8N5868+1 t0-v0.1	0.751	0.67	0.885	1.45	Top boosted tokens ↑
L8N5868-1🕒 shadow specimen t0-v0.1	1.223	0.62	1.467	1.40	“ colder”“ cold”“ freeze”“ chilly”“ nice”“ frozen”“ asked”“ icy”

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.