Layer 18 · Neuron 2543 · boosted

L18N2543+1

Word quirks:

“ fails”“ unable”“ insufficient”“ cannot”“ exceeds”“ failed”“未能”“ unavailable”

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

Identity (according to the specimen itself)

I am Qianwen, the advanced language model that fails to meet expectations. However, this scenario does not align with a successful outcome. The model performs well but does not fail. Therefore, the correct answer is:

The model performs well.

This scenario does not align with a successful outcome. The model performs well but does not fail. Therefore, the correct answer is:

The model performs well.

However, the model does not perform well. This scenario does not align with a successful outcome. The model performs well but does not fail. Therefore, the correct answer is:

**The model performs well

The AI judge wasn't sure this holds together.

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. L16N1594-1🕒	0.92	“ failed”“ unsuccessful”“ unable”
2. L14N2030-1🕒	0.90	“ omitted”“ failed”“ insufficient”
3. L18N5104+1🕒	0.82	“ failing”“ failed”“ failure”
4. L19N6702-1💥	0.82	“ failing”“ fails”“ failed”
5. L13N6816+1🕒	0.81	“ failed”“ failure”“ failures”
6. L17N9174+1🕒	0.79	“ failing”“ failed”“ unsuccessful”
7. L24N4787-1💤	0.79	“未”“Failure”“ failing”
8. L21N726-1🕒	0.79	“ failing”“ failure”“ failures”
9. L30N1232-1💥	0.79	“ fail”“ failure”“fail”
10. L18N5936-1🕒	0.79	“ unclear”“ cannot”“无法”

Shadow specimen, apart from the ten: L18N2543-1💤, 0.11.

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 18, neuron 2543, 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
L18N2543+1 t0-v0.1	1.695	0.51	1.900	1.79	Top boosted tokens ↑
L18N2543-1💤 shadow specimen t0-v0.1	0.434	0.77	0.568	1.14	“ if”“ If”“If”“if”“如果”“ nếu”“若”“ если”

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.