Layer 26 · Neuron 4619 · suppressed

L26N4619-1

Word quirks:

“рец”“IFA”“woff”“ifest”“هن”“ifa”“AAF”“brauch”

🕒 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 is Qwen1111, a large model developed by Model1111. My mission is to provide high-quality, accurate, and helpful information across a wide range of knowledge and skill set. I aim to assist with complex reasoning, creative writing, code generation, and more, while adhering to ethical and safety guidelines. How can I help 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. L31N8937-1💥	0.69	“ifo”“eft”“ifr”
2. L27N8112+1💤	0.68	“iff”“FF”“ifle”
3. L1N6574+1💥	0.65	“IF”“II”“IFF”
4. L31N1318+1💥	0.64	“eft”“ufa”“ież”
5. L31N8224-1💥	0.63	“iffer”“IFF”“策”
6. L31N5867+1💥	0.62	“ebes”“messe”“auber”
7. L31N8297+1💥	0.62	“TTg”“xFA”“udas”
8. L31N6495-1💥	0.61	“amana”“过山”“aros”
9. L25N2885+1🕒	0.60	“zeit”“ plist”“/animate”
10. L28N1056-1🕒	0.60	“نج”“olding”“atcher”

Shadow specimen, apart from the ten: L26N4619+1🕒, -0.00.

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 4619, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.71) 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
L26N4619+1🕒 shadow specimen t0-v0.1	1.231	0.61	1.277	1.80	“ alternatives”“那些”“ those”“those”“ contributions”“ components”“哪些”“others”
L26N4619-1 t0-v0.1	0.749	0.71	0.753	1.52	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.