Layer 7 · Neuron 1739 · boosted

L7N1739+1

Word quirks:

“ your”“ my”“⏎”“ myself”“ yours”“ you”“ yourself”“ I”

💤 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 an artificial intelligence, a large language model trained by Google. My "mission" isn't to perform a specific task like a human would (such as writing or speaking), but rather to process information, learn from data, and assist with problem-solving. I don't have personal goals or desires in the same way humans do. Instead, my purpose is to help you make informed decisions based on facts and evidence. How can I support you in achieving your objectives?

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. L5N5380+1🕒	0.98	“ my”“ myself”“ your”
2. L8N4915+1💤	0.96	“⏎”“ yours”“ your”
3. L13N7013-1🕒	0.96	“ yours”“ my”“ myself”
4. L6N5656-1🕒	0.96	“ YOU”“<\|im_end\|>”“ you”
5. L8N1138-1💤	0.95	“ your”“ new”“ my”
6. L8N7282+1💤	0.94	“ yourself”“ your”“ you”
7. L5N7559+1🕒	0.94	“ my”“ your”“ myself”
8. L6N2708-1💤	0.93	“ my”“ your”“ myself”
9. L14N746-1💤	0.93	“ my”“ you”“ myself”
10. L6N4116-1🕒	0.93	“ you”“ me”“ YOU”

Shadow specimen, apart from the ten: L7N1739-1🕒, -0.07.

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 7, neuron 1739, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, top-1 agreement 0.65) 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
L7N1739+1 t0-v0.1	0.961	0.65	1.135	1.33	Top boosted tokens ↑
L7N1739-1🕒 shadow specimen t0-v0.1	2.373	0.42	2.779	2.13	“ rating”“ ratings”“ Rating”“-rating”“Rating”“ stars”“ rated”“rating”

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.