Layer 14 · Neuron 1194 · suppressed

L14N1194-1

Word quirks:

“ --”“ -”“ __”“ @”“ ```”“ [[”“ -->”“ </”

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

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Alibaba Cloud's team of researchers. My primary goal is to assist users in completing tasks related to natural language processing and understanding. However, I also strive to provide a seamless user experience by addressing the needs of developers and users who seek help with various types of queries.

In addition to my core capabilities, I aim to support users in completing tasks related to natural language processing and understanding. To achieve this goal, I have integrated advanced features that enable me to provide accurate and reliable responses.

Overall, I believe that I can contribute to the development of a more efficient and

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. L7N105-1🕒	0.85	“<\|im_end\|>”“ ```”“.`”
2. L9N4418-1💤	0.85	“ -->”“ --”“<\|im_end\|>”
3. L11N1463+1💤	0.85	“ \"”“ --”“ __”
4. L6N4051+1🕒	0.84	“ \"”“ --”“ ____”
5. L11N5806-1💤	0.84	“ \"”“ --”“ ____”
6. L15N7431+1🕒	0.82	“ #”“ "”“ @”
7. L7N1653+1💤	0.82	“ --”“ ⏎”“ -”
8. L7N2674+1💤	0.82	“ ____”“ __”“ --”
9. L16N410+1🕒	0.81	“ --”“ ..."”“ ——”
10. L12N2260+1🕒	0.81	“ --”“ ____”“ -”

Shadow specimen, apart from the ten: L14N1194+1💤, 0.08.

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 14, neuron 1194, 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
L14N1194+1💤 shadow specimen t0-v0.1	0.951	0.62	1.191	1.71	“ actually”“⏎”“ detailed”“⏎ ⏎”“ specifics”“ referenced”“ directly”“⏎ ⏎”
L14N1194-1 t0-v0.1	1.084	0.59	1.370	1.25	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.