Layer 28 · Neuron 2700 · suppressed

L28N2700-1

Word quirks:

“栓”“ Opt”“le”“ Fit”“ropp”“拟合”“智”“Imp”

💤 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, the latest large language model developed by Alibaba's Ta Feng Advanced Industro & Optimization Research Team (通义实验室). My design focuses on providing powerful, practical, and ethical AI assistance across diverse scenarios.

My Mission

To empower users and organizations by:

Enhancing productivity: Streamening tasks like coding, data analysis, and creative writing.
Enabling deep reasoning: Solving complex problems in science, math, and logic.
Ensizing accessibility: Providing multilingual

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. The Misrememberer✅ L22N4602+1	0.68	“联”“疏”“ Styles”
2. L28N7782+1💤	0.65	“接”“老虎”“致”
3. L25N6663-1💤	0.65	“坡”“ifact”“ILLS”
4. L28N8611-1🕒	0.64	“灵”“靈”“韬”
5. L25N6750+1💤	0.64	“agin”“学法”“智”
6. L29N1091+1💤	0.64	“ Motiv”“ Neur”“足够”
7. L22N3951-1💤	0.63	“ Keber”“志”“发泡”
8. L26N2921-1🕒	0.63	“lish”“坚持”“ Diz”
9. L29N7902-1🕒	0.63	“ l”“hit”“共振”
10. L23N7730+1💤	0.62	“ypy”“幕”“伦”

Shadow specimen, apart from the ten: L28N2700+1🕒, 0.37.

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 28, neuron 2700, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.37, top-1 agreement 0.81) 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
L28N2700+1🕒 shadow specimen t0-v0.1	0.406	0.79	0.453	1.30	“척”“ vl”“剖”“tings”“oref”“(prom”“glich”“ prom”
L28N2700-1 t0-v0.1	0.368	0.81	0.397	1.52	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.