Layer 14 · Neuron 2853 · suppressed

L14N2853-1

Word quirks:

“ simulated”“ hypothetical”“ fictional”“ imagined”“ imaginary”“ special”“ metaphor”“ fict”

💤 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 information and support across a wide range of topics. If you have any questions or need assistance, feel free to ask!

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. L11N1602-1🕒	0.94	“ fictional”“ hypothetical”“ imagined”
2. L13N8875+1🕒	0.92	“ simulated”“ hypothetical”“ imaginary”
3. L9N1524+1🕒	0.91	“ simulated”“ imaginary”“ hypothetical”
4. L12N5977+1💤	0.91	“ fictional”“ imaginary”“ simulated”
5. L9N1143-1🕒	0.91	“ imaginary”“ imagined”“ hypothetical”
6. L14N9088-1🕒	0.90	“ imagined”“ hypothetical”“ imaginary”
7. L14N7051+1🕒	0.90	“ imaginary”“ represented”“ imagined”
8. L6N8691-1🕒	0.89	“ fictional”“ imaginary”“ fict”
9. L19N9153-1🕒	0.89	“ hypothetical”“ fictional”“ fict”
10. L13N7395-1🕒	0.89	“ imaginary”“ symbolic”“ metaphor”

Shadow specimen, apart from the ten: L14N2853+1🕒, 0.05.

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 14, neuron 2853, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.99, top-1 agreement 0.57) 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
L14N2853+1🕒 shadow specimen t0-v0.1	1.037	0.61	1.260	1.73	“<\|im_end\|>”“ Japanese”“ Japan”“ ND”“ ⏎”“ Korean”“ Ban”“ FW”
L14N2853-1 t0-v0.1	0.988	0.57	1.206	1.53	Top boosted tokens ↑

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.