Layer 16 · Neuron 3128 · suppressed

L16N3128-1

Word quirks:

“⏎”“:”“":”“⏎⏎”“”:”“:.”“':”“<|endoftext|>”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5
I am a large language model developed by Tongyi Lab. My primary purpose is to assist users in completing tasks across various domains:

General Q&A: Providing accurate information and reasoning
Creative Writing
Code Generation & Programming
Multilingual Support
Logical Reasoning
Visual Analysis
Role-playing
Document Processing
Autonomous Agent Planning
Multimodal Understanding
Real-time Web Search
Translation
**

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. L14N2735-1💤	0.95	“⏎”“⏎⏎”“”:”
2. L12N1748+1🕒	0.93	“⏎”“<\|im_end\|>”“:.”
3. L10N462-1🕒	0.93	“⏎⏎”“⏎”“.:”
4. L29N619+1💥	0.92	“⏎”“:”“⏎⏎”
5. L18N5223-1🕒	0.92	“⏎⏎”“:”“.”
6. L17N521+1🕒	0.91	“:”“：”“:*”
7. L15N4086+1🕒	0.91	“:”“⏎⏎”“⏎”
8. L14N7428-1💤	0.91	“⏎”“:”“”:”
9. L16N942+1🕒	0.91	“”:”“:”“":”
10. L12N117+1🕒	0.91	“.:”“”:”“’:”

Shadow specimen, apart from the ten: L16N3128+1🕒, 0.20.

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 3128, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.58, 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
L16N3128+1🕒 shadow specimen t0-v0.1	0.721	0.65	0.809	1.22	“”“�”“””“ __”“ same”“ ‘”“ “”“ also”
L16N3128-1 t0-v0.1	0.577	0.69	0.766	1.26	Top boosted tokens ↑

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.