Layer 29 · Neuron 616 · boosted

L29N616+1

Word quirks:

“段”“sez”“ساب”“端”“CELL”“φαλ”“席”“件”

🕒 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 Tongys (part of the Tongys family). My core function is to act as a versatile AI tool designed to assists users with a wide range of tasks by leveraging advanced capabilities in reasoning, coding, and multilingual communication.

My Missions:

Solving Complex Problems

I can reason step-by-step to tackle math, logic, or scientific challenges.
For example, if you need help with a calculus problem or a

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. L31N9060+1💥	0.82	“条”“段”“套”
2. L30N8475+1🕒	0.80	“场”“只”“条”
3. L31N4356-1💥	0.79	“条”“段”“部”
4. L24N4599+1💤	0.79	“段”“系”“期”
5. L31N2493+1💥	0.79	“层”“期”“段”
6. L29N7332+1🕒	0.79	“条”“片”“圈”
7. L27N5703-1💤	0.79	“奚”“加索”“ystem”
8. L26N549-1💤	0.78	“仙”“季”“席”
9. L27N6900+1💤	0.78	“契”“净额”“章”
10. L28N8108+1💤	0.78	“介”“面”“实拍”

Shadow specimen, apart from the ten: L29N616-1💤, -0.08.

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 616, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.52, top-1 agreement 0.77) 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
L29N616+1 t0-v0.1	0.516	0.77	0.595	1.85	Top boosted tokens ↑
L29N616-1💤 shadow specimen t0-v0.1	0.226	0.79	0.311	1.30	“ instances”“ pairs”“ organisations”“ Lines”“ combinations”“ Instances”“ Definitions”“ structures”

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.