Layer 13 · Neuron 8325 · boosted

L13N8325+1

Word quirks:

“ sometimes”“ simpler”“ actually”“ confusing”“ clearer”“ often”“ simple”“ similar”

🕒 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)

Hello! I am Qwen3.5, the latest large language model developed by the Tongyi Lab. My mission is to provide accurate, high-quality assistance across a wide range of tasks. Here's a breakdown of my capabilities:

Core Capabilities:

Advanced Reasoning: I excel at complex logical analysis, enabling me to solve intricate problems with precision.
Multilingual Support: I can fluently process and generate content in over 100 languages, ensuring global accessibility.
Contextual Understanding: I maintain clear and coherent communication, making me ideal for professional

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. L12N4181-1🕒	0.80	“ simple”“ specific”“ usually”
2. L13N7511-1💤	0.80	“ sometimes”“ simply”“ similar”
3. L13N9018-1🕒	0.79	“ simpler”“ basic”“ additionally”
4. L11N4120-1🕒	0.79	“ often”“ somewhat”“ clearer”
5. L13N4031-1💤	0.78	“ simple”“ somehow”“ similar”
6. L10N1897+1🕒	0.78	“ basically”“ clearer”“ clear”
7. L11N3683+1💤	0.78	“ actually”“ sometimes”“ similarly”
8. L12N5240-1🕒	0.78	“ similar”“ often”“ sometimes”
9. L12N875-1💤	0.78	“ actually”“ really”“ simple”
10. L17N4458-1💤	0.77	“ truly”“ already”“ actually”

Shadow specimen, apart from the ten: L13N8325-1💤, -0.18.

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 13, neuron 8325, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.63) 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
L13N8325+1 t0-v0.1	0.797	0.63	1.035	1.22	Top boosted tokens ↑
L13N8325-1💤 shadow specimen t0-v0.1	0.714	0.66	0.910	1.19	“!”“ choice”“ crafted”“!</”“!"”“!\”“⏎”“!");”

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.