Layer 5 · Neuron 6491 · suppressed

L5N6491-1

Word quirks:

“ reference”“ referenced”“ “”“ \"”“ description”“ ‘”“ refer”“ phrase”

💤 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 accurate, helpful, and unbiased information. My mission is to assist users in their research, learning, and problem-solving by offering clear explanations, examples, and guidance on a wide range of topics. I strive to be informative, engaging, and easy to understand, ensuring that my responses are well-structured and relevant to the user's needs.

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. L7N5672-1💤	0.93	“ “”“ referring”“ ‘”
2. L3N586-1🕒	0.91	“ ‘”“ “”“ reference”
3. L8N2231-1💤	0.91	“ ‘”“ --”“ reference”
4. L3N5911+1💤	0.91	“ \"”“ reference”“ name”
5. L5N2324-1🕒	0.91	“ ‘”“ \"”“ referenced”
6. L4N6405+1🕒	0.91	“ ‘”“ “”“ reference”
7. L5N5907-1🕒	0.90	“ name”“ referring”“ referenced”
8. L5N5604+1🕒	0.90	“ reference”“ referenced”“?\”
9. L3N651+1🕒	0.89	“ ‘”“ “”“ refer”
10. L4N6402+1🕒	0.89	“ ‘”“ reference”“ '”

Shadow specimen, apart from the ten: L5N6491+1🕒, -0.08.

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 6491, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L5N6491+1🕒 shadow specimen t0-v0.1	2.395	0.42	2.749	2.36	“ balance”“ balanced”“ balances”“ totals”“ accounting”“ Balance”“ calculation”“ accounted”
L5N6491-1 t0-v0.1	0.757	0.64	0.919	1.57	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.