Specimen ledger / L11N2093-1

Layer 11 · Neuron 2093 · suppressed

L11N2093-1

Word quirks:

“⏎”“⏎ ⏎”“⏎⏎”“ ⏎⏎”“⏎ ⏎”“ ⏎”“ ⏎ ⏎”“ <U+00A0>”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Alibaba Cloud. My mission is to provide helpful and accurate assistance in a wide range of tasks, from answering questions to generating creative content. I am designed to be versatile and efficient, ensuring that I can adapt to various needs across different domains.

If you have any specific questions about my capabilities or limitations, please let me know! I'm here to help. 😊

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. L9N2343-1💤	0.99	“⏎”“ ⏎”“ ⏎”
2. L15N7186+1💤	0.99	“⏎”“⏎⏎”“⏎ ⏎”
3. L15N4412+1💤	0.99	“⏎”“⏎⏎”“⏎ ⏎”
4. L14N4577+1💤	0.99	“⏎”“⏎⏎”“ ⏎”
5. L8N249+1💤	0.98	“⏎”“ ⏎”“⏎ ⏎”
6. L14N4510+1💤	0.98	“⏎⏎”“⏎”“ ⏎”
7. L16N4195+1💤	0.98	“⏎”“⏎⏎”“ ⏎”
8. L11N8011-1💤	0.98	“⏎”“⏎⏎”“ ⏎”
9. L9N107-1💤	0.98	“⏎”“⏎⏎”“ ⏎”
10. L9N557+1💤	0.98	“⏎”“ ⏎”“ ⏎”

Shadow specimen, apart from the ten: L11N2093+1💤, 0.15.

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 11, neuron 2093, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, 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
L11N2093+1💤 shadow specimen t0-v0.1	0.730	0.67	0.814	1.24	“ yes”“ ⏎”“ question”“ sometimes”“ exact”“ actually”“ questions”“ confusion”
L11N2093-1 t0-v0.1	0.889	0.65	1.177	1.32	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.