Layer 11 · Neuron 6103 · boosted

L11N6103+1

Word quirks:

“ focus”“ focusing”“ ⏎”“ focused”“ typically”“ ⏎”“ often”“ foc”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large-scale language model developed by Alibaba Cloud's Tongyi Lab. My primary mission is to process and understand natural language inputs, which allows me to generate coherent responses across a wide range of topics. While I do not have a specific "mission" in the sense of needing to solve a single problem, my capabilities enable me to:

Analyze complex queries: I can handle intricate questions that require synthesizing information from multiple domains or understanding nuanced relationships between different concepts.
Generate diverse outputs: My training allows me to produce varied text outputs, including code snippets

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. L15N6721-1💤	0.87	“ focus”“ info”“ usually”
2. L12N5283+1🕒	0.87	“ focused”“ normally”“ focus”
3. L13N4278-1🕒	0.86	“ focus”“ focused”“ focusing”
4. L11N3988-1💤	0.86	“ focus”“ foc”“ focusing”
5. L13N9031+1🕒	0.85	“ focus”“ focusing”“ now”
6. L11N2835-1🕒	0.85	“ focus”“ sometimes”“ foc”
7. L12N7637-1🕒	0.85	“ focus”“ focusing”“ Focus”
8. L14N5422+1💤	0.85	“ sometimes”“ **”“ foc”
9. L10N2932-1🕒	0.84	“ focus”“ ⏎⏎”“ focused”
10. L14N4267+1🕒	0.84	“ focus”“ ⏎”“ ⏎⏎”

Shadow specimen, apart from the ten: L11N6103-1🕒, 0.24.

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 6103, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, 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
L11N6103+1 t0-v0.1	0.857	0.65	1.026	1.25	Top boosted tokens ↑
L11N6103-1🕒 shadow specimen t0-v0.1	1.225	0.57	1.525	1.30	“ description”“ style”“ reflecting”“ align”“ now”“ customer”“ vibe”“ quirky”

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.