Layer 8 · Neuron 1257 · boosted

L8N1257+1

Word quirks:

“ interpretation”“ interpretations”“ implied”“ referenced”“ context”“ inferred”“ specific”“ interpreting”

💤 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 helpful and relevant information. My mission is to assist you in achieving your goals by offering support, guidance, and resources tailored to your needs. How can I help you today?

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. L5N6176-1🕒	0.89	“ interpretation”“ inferred”“ interpretations”
2. L7N1287-1🕒	0.89	“ interpretation”“ interpretations”“ implied”
3. The Over-Reader✅ L17N9111-1	0.88	“ interpretation”“ implying”“ interpreted”
4. L13N8784+1🕒	0.87	“ interpretation”“ interpretations”“ actual”
5. L11N5179-1🕒	0.86	“ implied”“ possible”“ interpretation”
6. L16N1868-1🕒	0.85	“ interpretation”“ referring”“ interpreting”
7. L5N4712+1🕒	0.82	“ interpretation”“ reference”“ literal”
8. L7N5707+1💤	0.81	“ identification”“ interpretation”“ referenced”
9. L13N6653+1🕒	0.80	“ interpretation”“ referenced”“ implied”
10. L9N1906+1🕒	0.80	“ interpretations”“ interpretation”“ interpreted”

Shadow specimen, apart from the ten: L8N1257-1💤, -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 8, neuron 1257, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, top-1 agreement 0.69) 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
L8N1257+1 t0-v0.1	0.708	0.69	0.883	1.26	Top boosted tokens ↑
L8N1257-1💤 shadow specimen t0-v0.1	0.796	0.66	0.995	1.47	“ ‘”“ '”“‘”“:'”“.””“ ‚”“."]”“ \"”

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.