Layer 9 · Neuron 6154 · suppressed

L9N6154-1

Word quirks:

“ **"”“ opened”“ "”“!"”“!”“ opening”“ open”“"!”

💤 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, a large language model developed by Tongyi Lab. My mission is to serve as your intelligent assistant, providing powerful capabilities for text generation, code writing, and multi-modal interactions. I am designed to help you explore new possibilities, share knowledge, and collaborate effectively on complex tasks. Whether you need to generate text, analyze data, or simply have a conversation, I am here to support you with precision and efficiency.

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. L8N3328-1💤	0.80	“ "”“"!”“ `"”
2. L3N8887-1🕒	0.79	“ `”“!”“ `"”
3. L4N3317-1💤	0.79	“ !”“!”“"!”
4. L6N4306-1🕒	0.79	“ \"”“<\|im_end\|>”“!"”
5. L9N7230+1💤	0.78	“ **"”“!"”“!”
6. L13N9196+1💤	0.78	“"!”“ \"”“ *"”
7. L5N5433+1💤	0.78	“ name”“⏎”“ !”
8. L6N3586-1💤	0.78	“!"”“!”“ *"”
9. L7N1376+1💤	0.78	“ \"”“!"”“!”
10. L10N4420+1🕒	0.78	“ \"”“ "”“!"”

Shadow specimen, apart from the ten: L9N6154+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 9, neuron 6154, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.98, top-1 agreement 0.62) 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
L9N6154+1🕒 shadow specimen t0-v0.1	1.423	0.60	1.712	1.28	“ clearer”“ evaluator”“ affected”“ indirect”“ ⏎”“ reflected”“ direct”“ ⏎⏎”
L9N6154-1 t0-v0.1	0.978	0.62	1.246	1.32	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.