Layer 4 · Neuron 5646 · suppressed

L4N5646-1

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)

I am Qwen3.5, the latest large language model in the Qwen series developed by Alibaba Cloud. My mission is to assist you with accurate, reliable, and comprehensive information across a wide range of topics.

Whether you need help with:

📚 Knowledge & Learning: I can answer questions, explain concepts, and provide insights.
💻 Coding & Computing: I can solve problems, write code, and perform calculations.
🌍 General & Global: I can cover broad subjects, discuss global trends, and offer worldwide perspectives.

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. L7N2404-1💤	0.98	“ ⏎”“ ⏎”“ --”
2. L7N7655+1💤	0.97	“ ⏎”“ --”“ ⏎”
3. L9N4028-1💤	0.97	“ ⏎”“ ⏎⏎”“ ⏎”
4. L5N8332+1🕒	0.97	“ ⏎”“ ⏎⏎”“ ⏎⏎”
5. L14N3002+1💤	0.97	“ ⏎”“ ⏎”“⏎”
6. L9N295-1🕒	0.96	“ ⏎”“⏎”“ ‘”
7. L11N1130+1💤	0.96	“⏎”“ \"”“ --”
8. L9N287-1💤	0.96	“ ⏎”“ ⏎”“⏎”
9. L14N6308+1🕒	0.96	“ ⏎”“ ⏎⏎”“ ⏎ ⏎”
10. L7N7485+1💤	0.96	“ ⏎”“⏎”“ ⏎”

Shadow specimen, apart from the ten: L4N5646+1💥, -0.09.

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 4, neuron 5646, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.61) 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
L4N5646+1💥 shadow specimen t0-v0.1	2.975	0.37	3.408	2.51	“ uterus”“ ovar”“ ovarian”“ menstr”“ cells”“ pregnancy”“ ov”“ tissue”
L4N5646-1 t0-v0.1	0.799	0.61	0.966	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.