Layer 14 · Neuron 5554 · boosted

L14N5554+1

Word quirks:

“（”“，”“在”“：”“使用”“。”“的”“与”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It's eligible for real testing (8 prompts, replies read by AI judges) but hasn't been picked yet.

Identity (according to the specimen itself)

你好！我是 Qwen3.5，作为通义千问系列中最新一代的模型，我的使命是提供高质量、多语言支持下的智能服务。以下是我的身份与能力概览：

我是 Qwen3.5（通义千问）

身份：Qwen3.5 是通义千问系列的最新版本，具备强大的逻辑推理、代码生成及多模态分析能力。
特点：
超长上下文窗口：支持处理超长的文档和对话记录。
高精度理解：在数学计算、自然科学

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. L13N5161-1🕒	1.00	“。”“：”“（”
2. L4N799-1💥	1.00	“在”“（”“，”
3. L8N5028+1🕒	1.00	“。”“，”“（”
4. L13N2416-1💥	0.98	“（”“与”“。”
5. L4N3369+1💥	0.98	“（”“，”“在”
6. L6N1391-1💥	0.98	“。”“，”“（”
7. L5N2094-1🕒	0.98	“（”“。”“，”
8. L4N7557-1🕒	0.98	“（”“在”“的”
9. L4N1657-1💥	0.98	“。”“，”“在”
10. L3N8716-1💥	0.98	“（”“，”“。”

Shadow specimen, apart from the ten: L14N5554-1🕒, -0.17.

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 14, neuron 5554, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.48, top-1 agreement 0.54) is inside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), but it has not been admitted: the band is sampled at random, round-robin across layers, with a quota of 300 neurons per layer. Being inside the band does not guarantee a place in this pass. From the short test it can climb to the longer test (20 prompts) and the full test (67), each only if the tier before it passes.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
L14N5554+1 t0-v0.1	1.482	0.54	1.716	1.57	Top boosted tokens ↑
L14N5554-1🕒 shadow specimen t0-v0.1	1.700	0.53	2.051	2.00	“?\”“?”“?!”“ better”“ empowerment”“?p”“ appropriate”“?]”

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.