Layer 11 · Neuron 16 · boosted

L11N16+1

Word quirks:

“<|im_end|>”“:\”“ ⏎⏎”“ ⏎”“:*”“...\”“\n”“):\”

💤 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 Qwen3.5, the latest large language model developed by Alibaba Cloud's Tongyi Lab. My mission is to provide accurate, helpful, and engaging responses that align with my capabilities as a state-of-the-art AI system. Whether you're looking for creative writing assistance, technical analysis, or general knowledge, I strive to deliver high-quality, reliable information that meets your needs.

I am designed to:

Understand complex queries across multiple domains
Generate coherent and contextually appropriate responses
Maintain accuracy and clarity in all interactions
Support both human and machine learning applications
Provide real-time, up

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. L10N1828+1💤	0.93	“ ⏎”“:\”“\n”
2. L2N7637-1🕒	0.93	“\n”“<\|im_end\|>”“⏎”
3. L12N6839-1💤	0.92	“<\|im_end\|>”“:\”“!\”
4. L5N921-1🕒	0.92	“⏎”“\n”“<\|im_end\|>”
5. L8N2862-1🕒	0.92	“\n”“<\|im_end\|>”“?\”
6. L3N8127-1🕒	0.92	“<\|im_end\|>”“⏎”“?\”
7. L3N6152+1🕒	0.92	“\n”“?\”“.\”
8. L5N2186-1💤	0.91	“<\|im_end\|>”“ ⏎”“ ⏎⏎”
9. L3N375+1🕒	0.91	“<\|im_end\|>”“⏎”“?\”
10. L3N7518+1🕒	0.91	“<\|im_end\|>”“\n”“ ⏎”

Shadow specimen, apart from the ten: L11N16-1🕒, 0.50.

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 11, neuron 16, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.57, 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
L11N16+1 t0-v0.1	0.574	0.69	0.735	1.26	Top boosted tokens ↑
L11N16-1🕒 shadow specimen t0-v0.1	0.882	0.62	1.132	1.29	“⏎”“⏎⏎”“.”“ ⏎”“⏎ ⏎”“ --”“⏎ ⏎”“ info”

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.