Layer 24 · Neuron 7034 · suppressed

L24N7034-1

Word quirks:

“ SIX”“PLACE”“Sek”“λευ”“基质”“ ше”“六种”“esel”

🕒 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 Q-Ze, a versatile AI assistant designed to help you navigate the digital world with confidence and clarity. My mission is to be your thought partner and problem solver—whether you need help with:

📥 Practical tasks: Summing up long articles, comparing products, or organizing your files
💬 Creative support: Crafting content, brainstorming ideas, or refining your voice
🔥 Real-time insights: Explaining complex topics (like quantum physics or climate models) in simple terms
🤖 Technical guidance: Debugging code, understanding APIs, or

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. L29N5772-1💤	0.68	“六”“六种”“六一”
2. L26N7535-1🕒	0.65	“ Six”“ SE”“六”
3. L23N2924-1🕒	0.65	“ six”“six”“六个”
4. L24N6970-1💤	0.65	“ Sloan”“ehren”“ Sext”
5. L28N1013+1💥	0.65	“ сем”“Six”“六”
6. L26N8764-1🕒	0.65	“十一”“六”“ bols”
7. L28N9166+1💤	0.64	“塞”“ral”“一把”
8. L27N3861-1💤	0.63	“ seen”“ Sixth”“六”
9. L27N8043-1🕒	0.63	“六”“ Six”“Six”
10. L23N1612-1💤	0.63	“沈”“色”“ Leop”

Shadow specimen, apart from the ten: L24N7034+1💤, 0.44.

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 24, neuron 7034, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.72) 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
L24N7034+1💤 shadow specimen t0-v0.1	0.756	0.70	0.793	1.21	“其二”“ Spin”“ELL”“pees”“/do”“SP”“ Indo”“ SP”
L24N7034-1 t0-v0.1	0.806	0.72	0.872	1.41	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.