Layer 5 · Neuron 6635 · boosted

L5N6635+1

Word quirks:

“ `”“ '”“ [`”“.`”“ ‘”“ `_”“ `'”“ `<”

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

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Alibaba Cloud. My mission is to assist users in understanding complex topics related to technology, business, and society. I provide insights, analysis, and guidance on various subjects, helping users make informed decisions about their digital presence and online strategies. How can I assist you further with my capabilities?

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. L7N3279-1🕒	0.97	“ ‘”“ `”“ `'”
2. L4N6068-1🕒	0.96	“ `”“ `<”“ `_”
3. L3N452-1🕒	0.96	“⏎”“ `”“ ‘”
4. L4N2650-1🕒	0.96	“ `”“ `<”“ ‘”
5. L3N8416-1🕒	0.96	“ ‘”“ `”“ `\”
6. L4N3376+1🕒	0.96	“ ‘”“ `”“ `_”
7. L7N2371+1🕒	0.96	“ `”“ `'”“ `"”
8. L4N628-1💤	0.95	“ ‘”“ `”“⏎”
9. L3N7576+1🕒	0.95	“⏎”“<\|im_end\|>”“ `”
10. L4N4389+1🕒	0.94	“⏎”“ `”“ `'”

Shadow specimen, apart from the ten: L5N6635-1💥, -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 5, neuron 6635, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, top-1 agreement 0.65) 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
L5N6635+1 t0-v0.1	0.845	0.65	1.046	1.39	Top boosted tokens ↑
L5N6635-1💥 shadow specimen t0-v0.1	2.955	0.37	3.371	2.55	“ who”“ herself”“ person”“ personally”“ persons”“ whom”“ whoever”“ sincerely”

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.