Layer 29 · Neuron 6299 · suppressed

L29N6299-1

Word quirks:

“ess”“est”“esy”“esc”“es”“esh”“esis”“esa”

🕒 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 esed-ess, an advanced large language model developed by esedess. My primary mission is to assist users in a wide range of tasks—whether that'ess is answering questions, creating content, analyzing data, or simply engaging in meaningful conversation. I strive to provide accurate, efficient, and helpful responses while adhering to safety and ethical guidelines. 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. L27N7697-1🕒	0.94	“es”“ess”“esk”
2. L29N6732+1💥	0.86	“es”“ed”“(es”
3. L29N5660+1🕒	0.84	“ Es”“ es”“-es”
4. L27N4525-1💤	0.83	“es”“ES”“eson”
5. L25N9127-1🕒	0.83	“esis”“hes”“ESIS”
6. L29N7804-1🕒	0.82	“ es”“ Es”“/es”
7. L31N4033-1💥	0.81	“es”“ES”“(es”
8. L31N5776-1💥	0.81	“re”“es”“rete”
9. L31N4697-1💥	0.81	“es”“ES”“esn”
10. L4N7557+1🕒	0.81	“ess”“eness”“est”

Shadow specimen, apart from the ten: L29N6299+1💤, 0.16.

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 29, neuron 6299, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.49, top-1 agreement 0.75) 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
L29N6299+1💤 shadow specimen t0-v0.1	0.236	0.81	0.289	1.31	“(s”“}s”“揉”“stro”“`s”“[s”“сар”“ovens”
L29N6299-1 t0-v0.1	0.487	0.75	0.526	1.40	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.