Specimen ledger / L12N2013-1

Layer 12 · Neuron 2013 · suppressed

L12N2013-1

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, a large language model developed by Alibaba Cloud's Tongyi Lab. My mission is to assist users in solving complex problems across various domains, including science, technology, and the arts. I aim to provide accurate, reliable, and helpful responses while maintaining high standards of safety and ethics.

I can help with tasks such as:

Writing stories or articles
Solving math or logic problems
Analyzing data
Generating code
Translating languages
And many more.

I strive to be a versatile tool for learning, creativity, and productivity. Whether you need

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. L9N22-1💤	0.97	“ ⏎”“ ⏎⏎”“ ⏎⏎”
2. L7N7485+1💤	0.97	“ ⏎”“⏎”“ ⏎”
3. L14N301+1💤	0.97	“ ⏎”“ ⏎”“ ⏎⏎”
4. L9N1363+1💤	0.97	“ ⏎”“ ⏎”“ --”
5. L14N6308+1🕒	0.97	“ ⏎”“ ⏎⏎”“ ⏎ ⏎”
6. L8N7653+1💤	0.97	“ ⏎”“ ⏎⏎”“ ⏎ ⏎”
7. L8N4658+1🕒	0.96	“ ⏎”“ ⏎⏎”“ ⏎”
8. L8N4007-1💤	0.96	“ ⏎”“ ⏎⏎”“ ⏎”
9. L14N1178+1💤	0.96	“ ⏎”“ ⏎⏎”“ ——”
10. L6N7284+1💤	0.96	“ ⏎”“⏎”“ —”

Shadow specimen, apart from the ten: L12N2013+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 12, neuron 2013, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.49, top-1 agreement 0.70) 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
L12N2013+1🕒 shadow specimen t0-v0.1	2.316	0.39	2.656	2.67	“ info”“ kids”“ ppl”“ basically”“ meds”“ help”“ pics”“ docs”
L12N2013-1 t0-v0.1	0.493	0.70	0.665	1.35	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.