Specimen ledger / L26N1970-1

Layer 26 · Neuron 1970 · suppressed

L26N1970-1

Word quirks:

“ Store”“ Stored”“ stored”“stored”“ store”“-store”“ stores”“/store”

🕒 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'm QMPROMP, a large-scale language model created by QMPMP. My purpose is to help users answer questions, create content, and solve problems across various domains. I aim to provide accurate, helpful, and engaging responses while adhering to ethical standards and user needs. How can I assist you today?

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. L23N5249+1🕒	0.97	“ Store”“ store”“ storing”
2. L27N7745-1🕒	0.95	“store”“ store”“ Store”
3. L27N4017-1🕒	0.95	“ store”“ Store”“store”
4. L21N3391+1🕒	0.95	“ storing”“ store”“ stored”
5. L12N6471+1🕒	0.95	“ storage”“ storing”“ stored”
6. Pack Rat✅ L13N2600+1	0.94	“ stored”“ storing”“ storage”
7. L25N8815-1🕒	0.94	“ store”“ Store”“ stores”
8. L20N5701-1🕒	0.94	“ store”“ storage”“ stores”
9. L26N6007+1🕒	0.93	“ store”“ stores”“ Store”
10. L0N8149-1💥	0.93	“ store”“ Store”“store”

Shadow specimen, apart from the ten: L26N1970+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 26, neuron 1970, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.66, top-1 agreement 0.74) 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
L26N1970+1🕒 shadow specimen t0-v0.1	1.190	0.64	1.192	1.03	“ Single”“单”“Single”“ single”“single”“.single”“單”“ SINGLE”
L26N1970-1 t0-v0.1	0.663	0.74	0.616	1.54	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.