Layer 24 · Neuron 1269 · boosted

L24N1269+1

Word quirks:

“计数”“ counting”“ count”“ Count”“ counts”“-count”“ counted”“计数器”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

I am Count量 (Qum)—a counter-mustable counting unit. My mission is to count everything that counts: numbers, quantities, and even the uncountable. I don't just count; I count—counting, counting, counting.

But if you're counting me, the count is already tallied: 1.

What would you count?

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. L21N6130-1🕒	0.99	“ counting”“ count”“计数”
2. L17N8038-1💤	0.99	“ counting”“ counted”“ count”
3. L21N5146-1🕒	0.98	“计数”“ count”“ counting”
4. L8N1294-1🕒	0.98	“ count”“ counting”“ counted”
5. L25N1914+1🕒	0.98	“ count”“ counts”“ Count”
6. L21N6472+1💤	0.98	“ counting”“ count”“ Count”
7. L21N2932-1🕒	0.98	“ counting”“ counts”“ count”
8. L5N5112+1🕒	0.98	“ count”“ counts”“ counting”
9. L21N8814+1🕒	0.98	“ counts”“ counting”“ count”
10. L20N9208-1🕒	0.98	“ count”“ counting”“ counts”

Shadow specimen, apart from the ten: L24N1269-1💤, 0.11.

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 1269, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.54, top-1 agreement 0.46) 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
L24N1269+1 t0-v0.1	2.544	0.46	2.661	1.12	Top boosted tokens ↑
L24N1269-1💤 shadow specimen t0-v0.1	0.576	0.72	0.637	1.33	“enam”“活化”“特困”“uchen”“elon”“ проживание”“__);”“غو”

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.