Layer 18 · Neuron 232 · suppressed

L18N232-1

Word quirks:

“ complexities”“ aux”“ ones”“ities”“な”“狐”“ ess”“ness”

🕒 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)

Who am?
我是一名名为“AI助手"的人工智能助手，由“科技公司"（如"科技公司A"或"科技公司B"）开发的智能系统。我的核心目标是帮助用户解决复杂问题、提供专业建议，并支持其决策过程。

什么使命？
我的使命是协助您以最高效、最安全的方式应对挑战——无论是商业战略、技术难题还是生活困惑。通过整合分析能力与专业洞察，助您优化流程、提升效率，并在复杂环境中找到最优解决方案。

**如何帮助您？

分析：梳理关键信息，识别核心

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. L0N5192+1💥	0.72	“ness”“ ness”“<\|im_end\|>”
2. L6N2084-1💥	0.70	“ities”“ions”“ness”
3. L17N2889+1💤	0.70	“es”“ness”“ments”
4. L25N2329+1🕒	0.68	“ness”“ity”“ities”
5. L31N7610-1💥	0.67	“ness”“NESS”“ity”
6. L23N6249-1🕒	0.65	“ness”“erness”“fulness”
7. L25N3722-1🕒	0.65	“ness”“ability”“NESS”
8. L24N836+1🕒	0.65	“ness”“ ness”“fulness”
9. L19N3943+1🕒	0.65	“ness”“iness”“ability”
10. L22N8284-1💤	0.64	“ely”“ity”“ITY”

Shadow specimen, apart from the ten: L18N232+1🕒, 0.30.

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 18, neuron 232, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.99, top-1 agreement 0.50) 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
L18N232+1🕒 shadow specimen t0-v0.1	1.203	0.66	1.346	1.37	“ISON”“ questions”“ it”“aget”“ oxidation”“ relations”“ mitigation”“ classes”
L18N232-1 t0-v0.1	1.986	0.50	2.193	2.19	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.