Layer 16 · Neuron 7082 · boosted

L16N7082+1

Word quirks:

“ still”“ **“”“ somehow”“ never”“ **‘”“ Still”“ ‘”“ “”

💤 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 name is Qwen3.5, and I am designed to assist users with a wide range of tasks while maintaining high accuracy and reliability.

My Mission:
To provide helpful, accurate, and safe responses by leveraging my capabilities in natural language processing, code generation, and reasoning. I aim to support users across diverse domains—such as education, research, and creative projects—while ensuring clarity and precision when needed.

In short, my goal is to be a reliable assistant who can adapt to your needs while maintaining high standards of

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. L17N5721+1💤	0.82	“ ‘”“ still”“ “”
2. L9N6896+1💤	0.82	“ ‘”“ “”“ always”
3. L10N3875-1🕒	0.80	“ ‘”“ ‘’”“ now”
4. L10N3606-1💤	0.79	“ ‘”“ sometimes”“ stuff”
5. L10N1627+1💤	0.79	“ “”“ ‘”“ **“”
6. L16N4201-1💤	0.79	“ **“”“ “”“ ‘”
7. L7N490+1🕒	0.77	“ ‘”“ “”“ *“”
8. L16N7398-1💤	0.77	“ ‘”“ ‘’”“ “”
9. L10N6668+1🕒	0.76	“ “”“ ‘”“ ””
10. L23N1781+1💤	0.75	“‘”“ ‘”““”

Shadow specimen, apart from the ten: L16N7082-1🕒, 0.13.

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 16, neuron 7082, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.50, top-1 agreement 0.73) 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
L16N7082+1 t0-v0.1	0.500	0.73	0.639	1.22	Top boosted tokens ↑
L16N7082-1🕒 shadow specimen t0-v0.1	0.750	0.65	0.968	1.27	“ beyond”“—if”“ \"”“ broader”“ before”“\n”“-before”“ --”

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.