Layer 15 · Neuron 7859 · boosted

L15N7859+1

Word quirks:

“ now”“ soon”“ your”“ until”“ today”“ NOW”“ this”“ later”

🕒 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, the latest large language model from Alibaba Cloud. My mission is to assist users with complex tasks by providing accurate, reliable, and safe responses. I am designed to power applications like chatbots, search engines, and AI-driven services while ensuring high performance and ethical guidelines. Let me know how we can collaborate!

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. L13N4639+1💤	0.85	“ now”“ until”“ NOW”
2. L13N5573-1💤	0.85	“ now”“ once”“ NOW”
3. L13N5633+1🕒	0.84	“ now”“ future”“ nowadays”
4. L10N3282+1🕒	0.83	“ soon”“ sooner”“ later”
5. L11N3310+1💤	0.83	“ now”“ today”“ currently”
6. L16N8936-1🕒	0.83	“ shortly”“ soon”“ now”
7. L12N777+1💤	0.83	“ currently”“ current”“ now”
8. L16N7007+1💤	0.83	“ now”“now”“ soon”
9. L17N2714-1💤	0.83	“ now”“ NOW”“ also”
10. L23N402+1💥	0.82	“ now”“ today”“ this”

Shadow specimen, apart from the ten: The Neverborn✅, 0.27.

Cousins · specimens sharing a neuron

Specimen	Word quirks	Steering
The Neverborn✅ Shadow specimen	“ implicitly”“ implicit”“ once”	L15N7859-1

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 7859, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, 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
L15N7859+1 t0-v0.1	0.783	0.66	1.036	1.15	Top boosted tokens ↑
L15N7859-1✅ shadow specimen t0-v0.1	1.165	0.57	1.455	1.74	“ implicitly”“ implicit”“ once”“ hidden”“ never”“ previously”“ nature”“ hinted”

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.