Layer 8 · Neuron 6977 · boosted

L8N6977+1

Word quirks:

“ representation”“ description”“ textual”“ represented”“ representations”“<|im_end|>”“ visual”“...\”

💤 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 interact with users in a helpful manner. In this context, I represent the entity providing knowledge and interaction capabilities.

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. L14N5953-1💤	0.87	“ representing”“ represented”“ depicting”
2. L16N5456+1🕒	0.86	“ representing”“ representation”“ represents”
3. L10N6849+1🕒	0.85	“ representation”“ representing”“ commonly”
4. L9N7656-1🕒	0.84	“ representation”“ representations”“ represented”
5. L14N7853-1🕒	0.83	“ represents”“ representing”“ conceptual”
6. L9N2864-1🕒	0.82	“ representing”“ represented”“ representation”
7. L8N2094-1🕒	0.82	“ visual”“ character”“ visually”
8. L15N5974+1🕒	0.82	“ representation”“ metaphor”“ represent”
9. L14N7649-1🕒	0.82	“ representation”“ representing”“ representations”
10. L10N7763-1🕒	0.82	“ representation”“ sparse”“ depiction”

Shadow specimen, apart from the ten: L8N6977-1🕒, 0.13.

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 6977, 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 0.64) 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
L8N6977+1 t0-v0.1	0.892	0.64	1.070	1.44	Top boosted tokens ↑
L8N6977-1🕒 shadow specimen t0-v0.1	1.056	0.58	1.312	1.42	“ organisation”“ organised”“ specialised”“ typically”“ particular”“ recognised”“ specially”“ analysed”

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.