Layer 2 · Neuron 4129 · suppressed

L2N4129-1

Word quirks:

“<|im_end|>”“ reciprocal”“ recipro”“ recip”“⏎”“ exchange”“ mutual”“ <->”

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

Identity (according to the specimen itself)

I am an AI assistant designed to provide information and support.

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. L4N7899+1💥	0.87	“ reciprocal”“ recipro”“ recip”
2. The Diplomat✅ L11N8437-1	0.85	“ recipro”“ recip”“ reciprocal”
3. L17N5139+1🕒	0.84	“ recipro”“ recip”“ reciprocal”
4. L2N5895+1💥	0.83	“ mutual”“<\|im_end\|>”“ recipro”
5. L1N6730+1💥	0.83	“ mutual”“ reciprocal”“<\|im_end\|>”
6. L10N8714+1🕒	0.79	“ mutual”“ recipro”“ reciprocal”
7. L1N5143+1💥	0.75	“<\|im_end\|>”“ <->”“⏎”
8. L3N4693-1💥	0.74	“ exchange”“ exchanged”“<\|im_end\|>”
9. L24N5506+1🕒	0.74	“ recip”“ reciprocal”“ recipro”
10. L17N5683-1🕒	0.73	“ interfering”“ exchanging”“ apart”

Shadow specimen, apart from the ten: L2N4129+1🕒, 0.06.

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 4129, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.85, top-1 agreement 0.34) 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
L2N4129+1🕒 shadow specimen t0-v0.1	2.307	0.42	2.736	2.57	“<\|im_end\|>”“ productions”“ production”“ Production”“⏎”“-production”“production”“ produced”
L2N4129-1 t0-v0.1	2.848	0.34	3.360	2.60	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.