Layer 3 · Neuron 4523 · boosted

L3N4523+1

Word quirks:

“!”“！”“<|im_end|>”“ ‘”“!**”“ !”“"!”“！**”

🕒 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 Qwen3.5，一个由阿里巴巴开发的大型语言模型。我的使命是协助你解答问题、提供信息或进行创作。无论是文本分析、代码生成、逻辑推理，还是简单的对话交流，我都在这里为你提供帮助。请告诉我具体需要帮助的内容，我会尽力满足你的需求！

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. L4N623+1🕒	0.98	“!”“ ‘”“!””
2. L4N7126-1🕒	0.98	“"!”“!"”“ ‘”
3. L3N2621-1🕒	0.97	“！”“!”“ ‘”
4. L3N8424+1🕒	0.96	“"!”“！”“!”
5. L5N8901+1🕒	0.96	“!”“"!”“ ‘”
6. L3N4385+1🕒	0.96	“<\|im_end\|>”“!"”“!**”
7. L16N5193-1🕒	0.96	“!”“!”“!”
8. L8N8424-1💤	0.95	“<\|im_end\|>”“!\”“!”
9. L5N2268-1💥	0.95	“!"”“!”“ !”
10. L4N8417+1💤	0.95	“ ‘”“!”“!””

Shadow specimen, apart from the ten: L3N4523-1💥, -0.04.

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 3, neuron 4523, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.51, top-1 agreement 0.53) 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
L3N4523+1 t0-v0.1	1.514	0.53	1.853	1.91	Top boosted tokens ↑
L3N4523-1💥 shadow specimen t0-v0.1	3.074	0.35	3.595	2.55	“<\|im_end\|>”“ straps”“ strap”“ attach”“ buttons”“ accessory”“ pins”“ accessories”

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.