Layer 11 · Neuron 3117 · boosted

L11N3117+1

Word quirks:

“ response”“ instantly”“ recalled”“ matched”“ immediate”“ basically”“ reflected”“ separate”

💤 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. My mission is to provide high-quality, accurate assistance by helping users get the most out of their interactions. I aim to be a reliable partner in delivering value, whether that's answering questions, solving problems, or simply offering a friendly conversation.

How can I help you today? If you'd like to continue the conversation, I can assist with tasks such as:

Answering factual questions based on my training data.
Providing creative ideas or brainstorming support for projects.
Offering step-by-step guidance on complex topics.
Or simply engaging in casual

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. L11N9161+1🕒	0.69	“ response”“ responses”“ instantly”
2. L12N1601-1🕒	0.66	“ related”“⏎”“ specific”
3. L11N6889-1💤	0.64	“ reflected”“ based”“ corresponding”
4. L9N8638-1💤	0.64	“ now”“ instantly”“ reflected”
5. L9N6067-1🕒	0.63	“ immediate”“ response”“ responding”
6. L11N6913+1🕒	0.63	“ response”“ reflected”“ earlier”
7. L5N3811-1💥	0.62	“ immediate”“ immedi”“ instantly”
8. L11N650+1🕒	0.62	“ reflected”“ instantly”“ earlier”
9. L13N2126-1🕒	0.62	“ first”“ same”“ quick”
10. L9N6741+1🕒	0.61	“ instantly”“ instant”“ instantaneous”

Shadow specimen, apart from the ten: L11N3117-1🕒, 0.25.

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 11, neuron 3117, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.69) 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
L11N3117+1 t0-v0.1	0.698	0.69	0.867	1.30	Top boosted tokens ↑
L11N3117-1🕒 shadow specimen t0-v0.1	1.286	0.59	1.505	1.55	“ usual”“ normally”“ typical”“ specially”“ related”“ extra”“ unspecified”“ usually”

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.