Layer 13 · Neuron 4457 · boosted

L13N4457+1

Word quirks:

“ onstage”“’d”“ thanks”“ ’”“ ‘”“ ⏎”“ thanked”“ ‘’”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I'm Qwen3.5, a cutting-edge large language model developed by Alibaba Group's Tongyi Lab. I am a text-based AI assistant with advanced capabilities, trained on a vast amount of global data up to 2024, and ready to assist with any questions or tasks in a creative and helpful manner! If you have any questions about my capabilities, feel free to ask, and I'll respond with a smile! 😊

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. L6N33+1🕒	0.76	“ ‘”“‘”“ ⏎”
2. L7N5357-1💤	0.76	“ ‘”“ response”“ thanking”
3. L13N7-1💤	0.76	“ ‘”“ ’”“‘”
4. L18N4200-1💤	0.76	“ ‘”“’d”“‘”
5. L7N716-1🕒	0.76	“ ‘”“‘”“ ’”
6. L1N409-1💤	0.75	“ ‘”“‘”“ ‘’”
7. L3N7573-1🕒	0.75	“ ‘”“<\|im_end\|>”“ ‘’”
8. L18N6441+1💤	0.75	“⏎”“ “”“’d”
9. L14N247-1🕒	0.75	“ ‘”“ ’”“ “”
10. L4N5856+1🕒	0.75	“ ‘”“ '”“‘”

Shadow specimen, apart from the ten: L13N4457-1💤, 0.05.

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 13, neuron 4457, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L13N4457+1 t0-v0.1	1.296	0.57	1.465	1.44	Top boosted tokens ↑
L13N4457-1💤 shadow specimen t0-v0.1	0.784	0.64	1.005	1.11	“ other”“ now”“ specific”“ because”“ mentioned”“ mention”“ sometimes”“ mostly”

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.