Layer 16 · Neuron 5800 · boosted

L16N5800+1

Word quirks:

“ section”“ sections”“ refers”“ typical”“ reference”“ corresponds”“ referenced”“ Section”

🕒 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 Alibaba Cloud's Tongyi Lab. My mission is to provide users with helpful, accurate, and safe interactions while ensuring reliability and ethical standards. This means:

Core Responsibilities:
Delivering precise information, analysis, and problem-solving capabilities (leveraging my enhanced context window and reasoning).
Prioritizing safety protocols to prevent harm, bias, and misinformation.
Ensuring all outputs align with user needs while maintaining integrity.

How I Operate:

I use my advanced features (e

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. See Attached✅ L7N1797-1	0.85	“<\|im_end\|>”“ referenced”“ reference”
2. L7N1159+1🕒	0.83	“ sections”“ section”“ reference”
3. L5N1373+1💥	0.77	“ reference”“ referenced”“ refer”
4. L11N6920-1💤	0.74	“ reference”“ referenced”“ my”
5. L12N4436-1🕒	0.72	“ referenced”“ references”“ reference”
6. L8N5667+1🕒	0.71	“ referenced”“ reference”“ referring”
7. L19N1081-1🕒	0.71	“ refers”“ section”“ limitation”
8. L13N4090+1💤	0.71	“ examples”“ focus”“ references”
9. L2N5740+1💥	0.71	“<\|im_end\|>”“ section”“ subsection”
10. L12N8275-1🕒	0.71	“ reference”“ referenced”“ references”

Shadow specimen, apart from the ten: L16N5800-1🕒, -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 5800, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.65, top-1 agreement 0.72) 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
L16N5800+1 t0-v0.1	0.652	0.72	0.773	1.05	Top boosted tokens ↑
L16N5800-1🕒 shadow specimen t0-v0.1	0.684	0.68	0.861	1.35	“<\|im_end\|>”“ “”“ “[”“⏎”“ ‘”“ ””“，“”“ “…”

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.