Layer 16 · Neuron 2333 · boosted

L16N2333+1

Word quirks:

“⏎”“ ⏎”“ ⏎”“ ⏎”“ ⏎⏎”“ everything”“ what”“ someone”

💤 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 developed by Alibaba Cloud. I am designed to provide high-quality, accurate information and support across a wide range of topics.

My Mission:
To empower individuals and organizations through knowledge and innovation. This means delivering precise, context-aware assistance that enhances productivity and learning. Whether in education, healthcare, or business, I aim to make complex tasks accessible while maintaining ethical standards.

Key Capabilities:

Precision: Delivering accurate information on technical and creative tasks.
Multilingual Support: Fluent communication in 100+ languages.
**Code &

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. L14N169+1💤	0.91	“⏎”“ ⏎”“ ⏎”
2. L15N3411-1💤	0.90	“⏎”“ ⏎”“ ⏎”
3. L8N275+1🕒	0.90	“ ⏎”“⏎”“ ⏎⏎”
4. L14N1381+1💤	0.90	“ ⏎”“⏎”“ ⏎ ⏎”
5. L14N775-1🕒	0.89	“ ⏎”“⏎”“ ⏎”
6. L15N634+1🕒	0.89	“⏎⏎”“ ⏎”“⏎”
7. L9N2115-1🕒	0.88	“ ⏎”“⏎”“ ⏎”
8. L9N3605+1💤	0.88	“ ⏎”“ ⏎”“⏎”
9. L11N5169+1🕒	0.88	“ ⏎”“⏎”“ ⏎”
10. L11N666+1💤	0.88	“ ⏎”“ ⏎”“ ⏎ ⏎”

Shadow specimen, apart from the ten: L16N2333-1🕒, 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 2333, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.62, top-1 agreement 0.71) 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
L16N2333+1 t0-v0.1	0.621	0.71	0.764	1.35	Top boosted tokens ↑
L16N2333-1🕒 shadow specimen t0-v0.1	1.324	0.54	1.612	2.32	“ your”“ you”“ this”“ that”“ simulator”“ what”“ Your”“ here”

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.