Layer 15 · Neuron 2639 · boosted

L15N2639+1

Word quirks:

“ previously”“ referenced”“ associated”“ referred”“ reflected”“ listed”“ mentioned”“ categorized”

🕒 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 Qwen3.2, a large language model. I assist users by generating text, analyzing data, and providing solutions across various domains. My capabilities include:

Code Generation & Debugging: Creating and refining code in multiple programming languages.
Data Analysis & Visualization: Processing structured and unstructured data to derive insights.
Multilingual Support: Communicating fluently in over 100 languages globally.
Long-context Understanding: Retaining information from lengthy documents or datasets.
Task Automation: Streamlining workflows through precise planning and execution.

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. L20N1276+1🕒	0.86	“ mentioned”“ corresponding”“ provided”
2. L18N4853+1💤	0.85	“ described”“ mentioned”“ referenced”
3. L13N1902+1🕒	0.85	“ referenced”“ referred”“ associated”
4. L15N8882+1💤	0.85	“ referenced”“ mentioned”“ noted”
5. L20N8986-1🕒	0.84	“ mentioned”“ discussed”“ described”
6. L13N5557+1💤	0.84	“ referenced”“ related”“ referred”
7. L13N6935+1💤	0.83	“ referenced”“ mentioned”“ covered”
8. L21N1468+1🕒	0.83	“ mentioned”“ charged”“-related”
9. L17N8161-1🕒	0.82	“ referenced”“ referred”“ mentioned”
10. L12N5392-1💤	0.82	“ mentioned”“ suggested”“ referenced”

Shadow specimen, apart from the ten: L15N2639-1🕒, -0.14.

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 15, neuron 2639, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, top-1 agreement 0.64) 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
L15N2639+1 t0-v0.1	0.851	0.64	1.103	1.31	Top boosted tokens ↑
L15N2639-1🕒 shadow specimen t0-v0.1	0.830	0.69	1.047	1.17	“ why”“ what”“ ______”“ ⏎”“ Why”“ example”“ ____”“ ⏎⏎”

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.