Layer 4 · Neuron 596 · boosted

L4N596+1

Word quirks:

“ ‘”“⏎”“‘”“ **“”“<|im_end|>”“ “”“ **‘”“ ——”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

我是 Qwen3.5，是阿里巴巴最新推出的通义千问大语言模型。我的使命是提供准确、安全、有益的信息和服务，同时遵循人类价值观和伦理准则。我致力于帮助用户解决问题、提供建议和支持，确保所有交互都是积极正面的。

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. L3N623+1💤	1.00	“ ‘”“‘”“ **‘”
2. L6N8825-1💤	0.98	“ ‘”“⏎”“<\|im_end\|>”
3. L5N3839-1🕒	0.98	“ ‘”“⏎”“ “”
4. L4N1216+1💤	0.98	“ ‘”“ “”“⏎”
5. L2N144-1🕒	0.98	“ ‘”“⏎”“ **“”
6. L3N931+1💤	0.98	“ ‘”“ “”“⏎”
7. L4N2669+1💤	0.98	“ ‘”“ “”“ ‘”
8. Prompt Echo✅ L4N7745-1	0.98	“ ‘”“ **‘”“<\|im_end\|>”
9. L4N2963-1🕒	0.97	“ ‘”“⏎”“‘”
10. L7N1253+1💤	0.97	“ ‘”“ ——”“⏎”

Shadow specimen, apart from the ten: L4N596-1💤, 0.55.

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 4, neuron 596, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, top-1 agreement 0.67) 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
L4N596+1 t0-v0.1	0.820	0.67	1.016	1.25	Top boosted tokens ↑
L4N596-1💤 shadow specimen t0-v0.1	0.900	0.62	1.106	1.63	“⏎”“<\|im_end\|>”“⏎⏎”“ ____”“ ______”“ _____”“ __”“⏎ ⏎”

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.