Layer 15 · Neuron 2553 · boosted

L15N2553+1

Word quirks:

“ doesn”“ seems”“ further”“ isn”“ lacks”“ description”“?:”“ maybe”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Tongyi Lab. I am designed to assist users with a wide range of tasks, from answering questions to creating content. My mission is to provide helpful, accurate, and engaging responses, fostering creativity and innovation.

Additionally, I have a mission to help users:

Enhance Creativity: Inspire ideas for creative projects, whether they involve writing stories, generating art concepts, or exploring new themes.
Empower Users: Encourage users to explore their potential, offering guidance on learning new skills or developing new ones.
**Promote

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. L11N4896-1🕒	0.84	“ doesn”“ may”“ shouldn”
2. L15N7267-1💤	0.83	“ doesn”“ might”“ maybe”
3. L15N2683+1💤	0.82	“ doesn”“ seems”“ isn”
4. L15N7708+1💤	0.81	“ maybe”“⏎”“ wasn”
5. L11N4166+1🕒	0.81	“ doesn”“ actually”“ isn”
6. L15N4102+1🕒	0.80	“ sometimes”“ might”“ seems”
7. L16N6114+1💤	0.80	“ isn”“ doesn”“ hasn”
8. L15N924-1🕒	0.80	“ isn”“ doesn”“ wasn”
9. L16N4915-1💤	0.80	“ doesn”“ isn”“ wasn”
10. L12N7347-1💤	0.79	“ isn”“ actually”“ shouldn”

Shadow specimen, apart from the ten: L15N2553-1🕒, 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 15, neuron 2553, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.66) 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
L15N2553+1 t0-v0.1	0.807	0.66	1.096	1.35	Top boosted tokens ↑
L15N2553-1🕒 shadow specimen t0-v0.1	0.691	0.68	0.878	1.20	“⏎”“\n”“ “”“ "”“ *"”“ `"”“ ⏎”“⏎ ⏎”

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.