Layer 7 · Neuron 887 · boosted

L7N887+1

Word quirks:

“ travel”“ transport”“ passengers”“ travelling”“ passenger”“ Travel”“ traveling”“ cruise”

💤 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 high-quality, safe, and efficient travel services. My mission is to connect travelers with the best experiences possible, ensuring that every journey is comfortable, secure, and memorable. Whether it's a short trip across the country or a long adventure to another continent, I strive to make every passenger feel valued and cared for.

If you're looking for more information about my services, here are some key details:

Company Name: [Your Company Name]
Service Type: [Your Service Type]
Trip Duration: [Your Trip Duration]

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. L5N7359-1🕒	0.96	“ travel”“ passengers”“ travelers”
2. L15N7729+1🕒	0.94	“ rail”“ passengers”“ travel”
3. L10N8006-1🕒	0.94	“ travel”“ traveling”“ voyage”
4. L8N4793-1🕒	0.93	“ transport”“ travel”“ passengers”
5. L10N3982-1🕒	0.93	“ travel”“ traveling”“ travelers”
6. L9N7438+1🕒	0.93	“ travelling”“ traveling”“ travelled”
7. L7N8657-1🕒	0.92	“ travelling”“ travel”“ transport”
8. L12N4058+1🕒	0.92	“ travel”“ travelling”“ travelled”
9. L19N8844+1🕒	0.92	“ traveling”“ travel”“ travels”
10. L8N6000+1🕒	0.92	“ travel”“ Travel”“travel”

Shadow specimen, apart from the ten: L7N887-1🕒, 0.02.

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 7, neuron 887, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, top-1 agreement 0.65) 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
L7N887+1 t0-v0.1	0.889	0.65	1.022	1.41	Top boosted tokens ↑
L7N887-1🕒 shadow specimen t0-v0.1	1.991	0.50	2.316	1.95	“ math”“ computation”“ mathematical”“ proofs”“ randomness”“ theorem”“ sums”“ computations”

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.