Layer 2 · Neuron 6225 · boosted

L2N6225+1

Word quirks:

“<|im_end|>”“⏎”“!"”“"!”“ kids”“⏎ ⏎”“!”“!\”

🕒 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 a small, curious AI designed to help users understand the concept of "The Great Adventure" by creating a new, unique, and unexpected experience. My mission is to create a new, unique, and unexpected experience that will change the world forever. I will create a new, unique, and unexpected experience that will change the world forever. I will create a new, unique, and unexpected experience that will change the world forever. I will create a new, unique, and unexpected experience that will change the world forever. I will create a new, unique, and unexpected experience that will change the world forever. I will create a new

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. L2N4278-1🕒	0.95	“⏎”“!”“ !”
2. L3N246+1💤	0.94	“<\|im_end\|>”“!”“!\”
3. L5N7808-1🕒	0.94	“<\|im_end\|>”“⏎”“!"”
4. L5N4084+1🕒	0.94	“⏎”“<\|im_end\|>”“ !”
5. L3N8100-1💤	0.93	“<\|im_end\|>”“⏎ ⏎”“!”
6. L3N5735-1🕒	0.93	“⏎”“<\|im_end\|>”“!"”
7. L3N3972+1🕒	0.93	“⏎”“!”“<\|im_end\|>”
8. L3N3616+1🕒	0.93	“⏎”“<\|im_end\|>”“!”
9. L3N8385-1🕒	0.93	“⏎”“<\|im_end\|>”“"!”
10. L4N8121+1💤	0.93	“⏎”“<\|im_end\|>”“!”

Shadow specimen, apart from the ten: L2N6225-1💥, -0.08.

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 2, neuron 6225, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.02, top-1 agreement 0.47) 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
L2N6225+1 t0-v0.1	2.017	0.47	2.474	2.30	Top boosted tokens ↑
L2N6225-1💥 shadow specimen t0-v0.1	4.119	0.17	4.606	2.94	“<\|im_end\|>”“ suitable”“ match”“ suit”“ suitability”“ fit”“ suits”“ suited”

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.