Layer 10 · Neuron 2577 · suppressed

L10N2577-1

Word quirks:

“ ONLY”“ --”“ NOT”“ now”“⏎⏎”“ not”“ only”“ ——”

💤 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 Alibaba Cloud. My mission is to provide high-quality assistance in areas such as coding, data analysis, and creative writing. If you have any questions or need help with tasks, feel free to let me know!

How can I assist 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. L10N3495-1🕒	0.78	“ ONLY”“ --”“⏎⏎”
2. L11N4269+1🕒	0.76	“ ONLY”“ only”“<\|im_end\|>”
3. L11N5428+1🕒	0.75	“ ONLY”“ only”“ NOT”
4. L7N7555+1🕒	0.73	“ --”“.--”“ ⏎ ⏎”
5. The Deflater✅ L15N8494+1	0.73	“ NOT”“ ONLY”“ only”
6. L10N7913+1🕒	0.73	“ ONLY”“ only”“ --”
7. L9N4699+1💤	0.73	“ only”“ sometimes”“ always”
8. L18N3513-1💤	0.72	“ not”“ only”“not”
9. L12N7400+1💤	0.72	“ ONLY”“ here”“ --”
10. L11N5631-1🕒	0.72	“ ONLY”“ only”“ NOT”

Shadow specimen, apart from the ten: L10N2577+1🕒, 0.11.

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 10, neuron 2577, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L10N2577+1🕒 shadow specimen t0-v0.1	1.924	0.46	2.323	2.06	“ colour”“⏎”“ colours”“ ____”“ ?”“⏎ ⏎”“ organisation”“ coloured”
L10N2577-1 t0-v0.1	0.763	0.65	0.947	1.31	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.