Layer 28 · Neuron 2477 · suppressed

L28N2477-1

Word quirks:

“ herself”“ eye”“身心”“心智”“nose”“眼”“ Eye”“ mind”

💤 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 Tongdun (Tongwan). I am designed to be a versatile and powerful AI assistant capable of handling complex tasks across various domains.

Who I Am

Identity: A state-of-the-art large language model with enhanced capabilities in reasoning, coding, multilingual support, and long-context understanding.
Core Strengths:
Advanced Reasoning: Solves math, science, and logic problems with step-by-step precision.
Full-stack Coding: Generates, debugs, and optim

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. L25N924-1🕒	0.88	“ eyes”“ eye”“ face”
2. Mindworm✅ L28N3090+1	0.81	“ mind”“ eye”“mind”
3. L24N5396-1🕒	0.78	“眼”“眼中”“心中”
4. L29N2967+1🕒	0.78	“ eye”“眼”“eye”
5. L30N691+1💥	0.76	“眼”“ eye”“ Eye”
6. L31N2800+1💥	0.76	“ mind”“眼”“ eye”
7. L24N5526+1🕒	0.75	“ eye”“眼的”“眼”
8. L29N5349+1🕒	0.75	“Eye”“eye”“聋”
9. L22N4315-1💤	0.75	“Eye”“ eye”“-eye”
10. L27N8749+1💤	0.74	“ eyes”“眼睛”“eyes”

Shadow specimen, apart from the ten: L28N2477+1🕒, -0.21.

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 2477, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.38, top-1 agreement 0.79) 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
L28N2477+1🕒 shadow specimen t0-v0.1	0.672	0.76	0.727	2.20	“.'”“_'”“>'”“}'”“*'”“.'"”“+'”“:'”
L28N2477-1 t0-v0.1	0.377	0.79	0.527	1.35	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.