Layer 5 · Neuron 3044 · boosted

L5N3044+1

Word quirks:

“ temperature”“ temperatures”“ range”“ ranges”“ Temperature”“ therm”“ thermal”“temperature”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

I am ready to process the input range $0.1 \leq x \leq 0.9$. My mission involves handling values within this interval, which allows for a wide dynamic range of operation.

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. L19N5696+1🕒	0.94	“ temperature”“ thermal”“ Temperature”
2. L7N3435-1🕒	0.92	“ temperature”“ thermal”“ heat”
3. L8N7128-1🕒	0.92	“ temperature”“ temperatures”“ Temperature”
4. L8N7358-1🕒	0.92	“ temperature”“ temperatures”“ Temperature”
5. L7N5369+1🕒	0.91	“ temperature”“ thermometer”“ therm”
6. L8N5207-1🕒	0.91	“ temperature”“ temperatures”“ Temperature”
7. L5N6983-1🕒	0.91	“ temperature”“ Temperature”“温度”
8. L7N7047-1🕒	0.91	“ temperature”“ heat”“ thermal”
9. L18N5656+1🕒	0.91	“ temperature”“ temperatures”“ Temperature”
10. L7N1514+1🕒	0.90	“ temperature”“ heat”“ temperatures”

Shadow specimen, apart from the ten: L5N3044-1🕒, -0.03.

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 5, neuron 3044, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.75, top-1 agreement 0.38) 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
L5N3044+1 t0-v0.1	2.748	0.38	3.170	2.43	Top boosted tokens ↑
L5N3044-1🕒 shadow specimen t0-v0.1	1.125	0.60	1.277	1.58	“ \"”“ processed”“ “”“ processing”“ reason”“ referenced”“ reference”“ generated”

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.