Layer 14 · Neuron 1368 · suppressed

L14N1368-1

Word quirks:

“ again”“ additionally”“ combined”“ together”“ also”“ additional”“ paired”“ partnered”

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

Identity (according to the specimen itself)

I am Qwen3.5, and I am also part of the broader ecosystem of AI models that contribute to a future where technology serves as a bridge between human creativity and digital intelligence.

I hope you find this information helpful! Let me know if you need any further details about my capabilities.

Enjoy your day! 😊

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. L14N3517+1🕒	0.86	“ combined”“ together”“ paired”
2. L13N6094+1🕒	0.84	“ plus”“ additionally”“ partners”
3. L17N5943+1💤	0.83	“ additionally”“ supplemented”“ paired”
4. L14N8948+1🕒	0.82	“ now”“ combined”“ also”
5. L15N5515+1💤	0.82	“ AND”“ combined”“ actually”
6. L13N2111+1🕒	0.82	“ additionally”“ combination”“ additional”
7. L16N8642+1🕒	0.82	“ plus”“ additionally”“ and”
8. L15N4612-1🕒	0.82	“ combined”“ plus”“ mixed”
9. L15N7827-1🕒	0.82	“ combined”“ plus”“ Combined”
10. L15N7365+1🕒	0.81	“ plus”“ additionally”“ combined”

Shadow specimen, apart from the ten: L14N1368+1💤, -0.13.

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 1368, 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
L14N1368+1💤 shadow specimen t0-v0.1	0.810	0.65	1.020	1.41	“⏎”“ ⏎”“ ‘’”“ __”“ ””“ ____”“ ‘”“ _____”
L14N1368-1 t0-v0.1	1.094	0.55	1.433	1.70	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.