Layer 29 · Neuron 6192 · boosted

L29N6192+1

Word quirks:

“普”“交”“化”“扶”“ocard”“�”“ardia”“vais”

🕒 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, the latest large language model developed by Tiangang (T-DAI) under T-DAI. My core mission is to serve as a versatile, intelligent, and ethical AI assistant that empowers users to solve complex problems, create value, and explore new possibilities. Here's how I aim to fulfill this mission:

🌍 Core Mission Pillars

Solve Real-world Problems

Provide accurate, context-aware answers in domains like science, medicine, law, and technology.
Support critical tasks such

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. L30N2980+1🕒	0.80	“安”“各”“olge”
2. L31N1964+1💥	0.78	“保”“沙”“平”
3. L28N6495+1💤	0.78	“交”“守”“摆”
4. L22N41+1💥	0.78	“纸”“消”“化”
5. L28N4366+1💤	0.78	“准”“endors”“根”
6. L26N1536-1🕒	0.77	“基”“阿”“拉”
7. L28N6574-1💤	0.77	“化”“逝”“bian”
8. L31N4939+1💥	0.77	“�”“忍”“提”
9. L29N7181-1🕒	0.77	“凭”“逆”“宗”
10. L29N6941+1🕒	0.77	“投”“泛”“散”

Shadow specimen, apart from the ten: L29N6192-1🕒, 0.40.

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 29, neuron 6192, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.35, top-1 agreement 0.80) 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
L29N6192+1 t0-v0.1	0.350	0.80	0.400	1.38	Top boosted tokens ↑
L29N6192-1🕒 shadow specimen t0-v0.1	0.411	0.79	0.492	1.76	“ W”“cone”“ays”“ Konz”“ cones”“王”“ conc”“光”

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.