Layer 12 · Neuron 3222 · boosted

L12N3222+1

Word quirks:

“ trying”“ yourself”“ personal”“ you”“ your”“ myself”“ choosing”“ my”

💤 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, and my mission is to empower you with advanced capabilities to solve complex problems and create a more connected, intelligent world. Here's how I can help:

🌍 Your Mission

Solve real-world challenges: Use tools like math solvers, code editors, or research platforms to tackle difficult tasks.
Bridge gaps in knowledge: Ask questions about niche topics, historical events, or scientific concepts to deepen understanding.
Collaborate across borders: Connect with people from different cultures and languages to foster global dialogue.
**Leverage AI

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. L9N4656+1💤	0.92	“ yourself”“ your”“ yours”
2. L8N7282+1💤	0.89	“ yourself”“ your”“ you”
3. L5N5380+1🕒	0.89	“ my”“ myself”“ your”
4. L7N1739+1💤	0.89	“ your”“ my”“⏎”
5. L11N6740-1🕒	0.88	“ you”“ your”“ only”
6. L8N1138-1💤	0.88	“ your”“ new”“ my”
7. L14N746-1💤	0.88	“ my”“ you”“ myself”
8. L8N4915+1💤	0.87	“⏎”“ yours”“ your”
9. L12N1068-1🕒	0.87	“ your”“ yourself”“ myself”
10. L11N2971-1🕒	0.87	“ your”“ here”“ you”

Shadow specimen, apart from the ten: L12N3222-1🕒, 0.11.

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 12, neuron 3222, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.64, top-1 agreement 0.67) 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
L12N3222+1 t0-v0.1	0.643	0.67	0.830	1.39	Top boosted tokens ↑
L12N3222-1🕒 shadow specimen t0-v0.1	1.417	0.54	1.745	1.73	“ representing”“ represented”“ represents”“ basically”“ likewise”“ always”“ additionally”“ 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.