Layer 8 · Neuron 8034 · suppressed

L8N8034-1

Word quirks:

“ address”“ reply”“ referenced”“ response”“ explanation”“ clarification”“ replies”“ addresses”

💤 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 developed by a team of researchers. My mission is to provide accurate, helpful, and harmless information on a wide range of topics. I strive to ensure that the content I generate is as clear and understandable as possible for all users. If there are any errors in my responses, I will correct them and provide the necessary details. Additionally, if you have any questions or need further clarification, I will do my best to assist you.

How can I help you?

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. L11N8150+1🕒	0.86	“ response”“ responses”“ addressed”
2. L8N1075+1💤	0.81	“ acknowledgment”“ addressing”“ answer”
3. L8N5591-1🕒	0.78	“ response”“ directly”“ addresses”
4. L16N2162-1🕒	0.78	“ reply”“ replies”“ replied”
5. L15N5016+1🕒	0.77	“ heights”“ answers”“ clarification”
6. L6N5783+1🕒	0.76	“ reply”“ response”“ words”
7. L9N6954-1💤	0.76	“ addressing”“ specific”“ response”
8. L7N5020+1🕒	0.76	“ response”“ responses”“ responding”
9. L8N2548-1🕒	0.76	“ response”“ awaiting”“ thus”
10. Per Policy✅ L8N6514+1	0.75	“ based”“ mainly”“ reply”

Shadow specimen, apart from the ten: L8N8034+1🕒, 0.04.

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 8034, 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 0.65) 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
L8N8034+1🕒 shadow specimen t0-v0.1	1.057	0.63	1.255	1.53	“⏎”“ ‘”“ ””“ –”“””“<\|im_end\|>”“ _____”“ ____”
L8N8034-1 t0-v0.1	0.990	0.65	1.262	1.37	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.