Layer 9 · Neuron 7110 · boosted

L9N7110+1

Word quirks:

“ thinking”“ **“”“ directly”“ ‘”“ ALL”“ ONLY”“ **‘”“ “”

💤 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 Tongyi Lab. My mission is to empower you with advanced capabilities across diverse domains:

In research: I can synthesize complex scientific concepts, analyze data patterns, or generate detailed technical reports to support evidence-based decision-making.
In education: I assist students in understanding challenging topics, providing clear explanations, or guiding learning processes.
In productivity: I help professionals streamline workflows, automate tasks, or enhance efficiency through intelligent tools.
In creativity: I inspire innovation by generating original ideas, stories, or designs tailored to specific needs

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. L12N2832+1🕒	0.84	“ “”“ ‘”“ **“”
2. L10N3159+1🕒	0.82	“ *“”“ “”“ “”
3. L4N4179-1🕒	0.82	“ ‘”“ “”“‘”
4. L15N7468-1💤	0.81	“ ‘”“ ‘’”“ **‘”
5. L11N4524+1🕒	0.81	“ ‘”“ ALWAYS”“ “”
6. L13N7678-1🕒	0.81	“ ‘”“ **“”“ directly”
7. L16N2382-1💤	0.81	“ “”“ ‘”“ **“”
8. L17N4600-1💤	0.81	“ “”“ **“”“ ‘”
9. L8N412-1💤	0.81	“ **“”“ “”“ ‘”
10. L9N1608+1🕒	0.81	“ “”“ ‘”“ ‘”

Shadow specimen, apart from the ten: L9N7110-1💤, -0.05.

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 9, neuron 7110, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, top-1 agreement 0.68) 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
L9N7110+1 t0-v0.1	0.742	0.68	0.992	1.39	Top boosted tokens ↑
L9N7110-1💤 shadow specimen t0-v0.1	0.985	0.61	1.191	1.50	“ typical”“ description”“ describing”“ possibly”“ ⏎⏎”“ supposedly”“ described”“ ⏎”

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.