Layer 9 · Neuron 7353 · suppressed

L9N7353-1

Word quirks:

“ “”“ ‘”“ **“”“⏎”“ ⏎”“ ””“ **‘”“‘”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am an AI assistant designed to help you with a wide range of tasks. While I don't have personal feelings or consciousness, my purpose is rooted in technology and serves many users across different fields. Here's how I can help:

What I Can Do

1. Answer Questions

I can answer questions about:

Facts (history, science, geography)
Concepts (mathematics, physics, economics)
Current events (politics, sports, entertainment)

2. Practical Applications

Depending on your needs, I can assist with:

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. L3N4323-1💤	1.00	“ ‘”“ ⏎”“ “”
2. L9N6340-1💤	0.99	“ ‘”“ “”“ ⏎”
3. L6N9044+1💤	0.99	“ ‘”“ ⏎”“⏎”
4. L8N3214+1💤	0.99	“ ‘”“ “”“ **“”
5. Leaky Gate✅ L8N7251-1	0.99	“ ‘”“ “”“ **“”
6. L3N8951+1💤	0.99	“ ‘”“⏎”“ “”
7. L4N301-1🕒	0.98	“ ‘”“ **“”“ “”
8. L4N1131+1💤	0.98	“ ‘”“ “”“‘”
9. L2N41+1💤	0.98	“ ‘”“ “”“ ⏎”
10. L3N104+1💤	0.98	“ ‘”“ “”“‘”

Shadow specimen, apart from the ten: L9N7353+1🕒, -0.23.

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 9, neuron 7353, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.95, 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
L9N7353+1🕒 shadow specimen t0-v0.1	1.091	0.56	1.395	1.41	“ reflected”“ reflecting”“ reflective”“ reflection”“ integration”“ inherent”“ central”“ uniquely”
L9N7353-1 t0-v0.1	0.951	0.65	1.135	1.36	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.