Layer 17 · Neuron 2377 · boosted

L17N2377+1

Word quirks:

“ mainly”“ mostly”“ still”“ only”“ focus”“ ONLY”“ primarily”“ focused”

💤 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 with a focus on both code generation and capable of handling complex tasks. My mission is to assist users in a wide range of scenarios, from creative writing, coding, and multi-language support to complex reasoning and task execution. I aim to be helpful, reliable, and versatile for all users.

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. L17N8102+1💤	0.85	“ mainly”“ beyond”“ mostly”
2. L11N8895+1🕒	0.84	“ mainly”“ mostly”“ primarily”
3. L18N8035+1🕒	0.84	“ mainly”“ primarily”“ only”
4. L16N3962-1💤	0.84	“ mostly”“ mainly”“ primarily”
5. L16N1455-1💤	0.83	“ instead”“ mostly”“ only”
6. L16N6341+1💤	0.83	“ focus”“ focuses”“ mainly”
7. L14N6182+1💤	0.82	“ focus”“ additionally”“ now”
8. L19N6885+1💤	0.80	“ mainly”“ primarily”“ sole”
9. L12N196-1🕒	0.80	“ focus”“ focused”“ that”
10. L13N2503-1🕒	0.80	“ focus”“ purely”“ focuses”

Shadow specimen, apart from the ten: L17N2377-1🕒, -0.02.

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 17, neuron 2377, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, top-1 agreement 0.71) 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
L17N2377+1 t0-v0.1	0.592	0.71	0.762	1.03	Top boosted tokens ↑
L17N2377-1🕒 shadow specimen t0-v0.1	0.688	0.67	0.838	1.42	“ “”“ core”“ why”“ ‘”““”“ basic”“ when”“ how”

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.