Layer 8 · Neuron 8003 · suppressed

L8N8003-1

Word quirks:

“ usually”“ often”“ typically”“ referenced”“ generally”“ specific”“ referring”“ sometimes”

💤 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 high-quality, helpful, and harmless responses. My mission is to empower users with accurate information, actionable advice, and creative solutions to their challenges.

Here is how I typically structure my responses:

Role: AI Assistant / Digital Companion
Mission: To offer clear, concise, and direct answers to user questions or problems.
Tone: Professional, objective, and supportive.

---

Example Response (Short & Direct)

Question: "How can I improve my productivity?"

Answer:
"I recommend starting by identifying

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. L12N5503-1🕒	0.97	“ usually”“ generally”“ typically”
2. L8N5674-1💤	0.96	“ typically”“ usually”“ typical”
3. L13N8424+1💤	0.94	“ usually”“ often”“ typically”
4. L11N6822+1🕒	0.94	“ referred”“ usually”“ often”
5. L17N3884+1💤	0.93	“ usually”“ typically”“ often”
6. L9N5402-1💤	0.93	“ usually”“ typically”“ generally”
7. L8N4971-1💤	0.93	“ specific”“ usually”“ typically”
8. L5N7890+1💤	0.92	“ usually”“ often”“ referring”
9. L10N8872-1🕒	0.92	“ sometimes”“ known”“ often”
10. L13N1027-1💤	0.92	“ referenced”“ typically”“ generally”

Shadow specimen, apart from the ten: L8N8003+1💤, -0.10.

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 8003, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, 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
L8N8003+1💤 shadow specimen t0-v0.1	0.914	0.59	1.137	1.22	“ ⏎”“⏎”“ ⏎⏎”“ ⏎”“<\|im_end\|>”“?\”“ ____”“ _____”
L8N8003-1 t0-v0.1	0.728	0.69	0.934	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.