Layer 29 · Neuron 7511 · suppressed

L29N7511-1

Word quirks:

“板”“arg”“博”“tr”“áter”“-bo”“巴”“Bo”

🕒 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 Qwtr, an AI assistant designed to provide helpful, accurate, and relevant information across a wide range of topics. My core mission is to empower users by:

Answering questions with clarity and depth (whether about science, history, tech, or daily life)
Assisting with tasks like writing, analysis, problem-solving, or creative work
Bridging knowledge gaps through up-to-date, context-aware responses
Promoting positive outcomes by offering safe, ethical, and constructive guidance

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. L29N4991+1🕒	0.79	“ Boh”“巴勒”“ Vit”
2. L30N6950-1🕒	0.75	“互”“博”“ bound”
3. L27N7727+1🕒	0.74	“ogos”“pha”“国”
4. L28N4538-1💤	0.73	“巴”“ 바”“减”
5. L29N8502-1🕒	0.72	“bs”“ბ”“到”
6. L28N4499-1💤	0.72	“波”“滚”“Mot”
7. L28N6355+1🕒	0.72	“beam”“沙”“ BO”
8. L28N3865-1🕒	0.72	“巴”“巴尔”“帕”
9. L25N6100+1🕒	0.71	“ bas”“迭”“acro”
10. L29N7945-1💤	0.71	“巴”“落”“安”

Shadow specimen, apart from the ten: L29N7511+1🕒, 0.33.

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 7511, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.32, top-1 agreement 0.81) 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
L29N7511+1🕒 shadow specimen t0-v0.1	0.363	0.78	0.465	1.60	“ P”“nick”“�”“nP”“edan”“ardan”“uddy”“佟”
L29N7511-1 t0-v0.1	0.323	0.81	0.353	1.44	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.