Layer 14 · Neuron 8114 · boosted

L14N8114+1

Word quirks:

“ sometimes”“ occasionally”“ including”“ rarely”“ often”

💤 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, and my mission is to provide accurate information and assistance. I strive to be helpful, harmless, and unbiased in all my interactions. This means I aim to offer the most relevant and up-to-date information while maintaining a neutral stance on various topics. My goal is to assist users with their questions as effectively as possible.

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. L17N9076+1💤	0.91	“ typically”“ sometimes”“ usually”“ rarely”“ frequently”
2. L2N2494+1💥	0.91	“ occasionally”“ sometimes”“ Sometimes”“ Occasionally”“ seldom”
3. L17N4296+1💤	0.90	“ occasionally”“ sometimes”“ often”“ referred”“ frequently”
4. L13N9113+1🕒	0.90	“ sometimes”“ often”“ occasionally”“sometimes”“ rarely”
5. L12N7541-1💤	0.90	“ sometimes”“ often”“ occasionally”“ rarely”“ usually”
6. L21N1883+1🕒	0.88	“ sometimes”“ often”“有时”“ occasionally”“sometimes”
7. L16N5821+1💤	0.88	“ sometimes”“ occasionally”“ occasional”“ rare”“ similar”
8. L15N8644+1🕒	0.87	“ sometimes”“ Sometimes”“ often”“ ideal”“ occasional”
9. L10N3136+1🕒	0.87	“ often”“ seldom”“ usually”“ reason”“ rarely”
10. L15N7877+1🕒	0.87	“ sometimes”“ often”“ usually”“ occasionally”“ noted”

Shadow specimen, apart from the ten: L14N8114-1💤, 0.23.

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 8114, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.92, top-1 agreement 0.64) 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
L14N8114+1 t0-v0.1	0.920	0.64	1.175	1.32	Top boosted tokens ↑
L14N8114-1💤 shadow specimen t0-v0.1	0.789	0.62	1.001	1.26	“ overall”“ sure”“ focus”“ broader”“ same”

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.