Layer 3 · Neuron 1506 · boosted

L3N1506+1

Word quirks:

“ triangle”“ angle”“ triangles”“ angles”“ Angle”“ Triangle”“ congr”“ perpendicular”

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

Identity (according to the specimen itself)

You've got ∠ABC. You want to find the measure of ∠ABC.
∠ABC = 180° - ∠DEF
∠DEF = 90° - ∠GHI
∠GHI = 45° - ∠JKL
∠JKL = 30° - ∠MNO
∠MNO = 60° - ∠PQR
∠PQR = 120° - ∠QRS
∠QRS = 135° - ∠STU
∠STU =

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. L4N3191+1💥	0.93	“ triangle”“ triangles”“ angle”
2. L3N1069-1💥	0.90	“ triangles”“ triangle”“ angle”
3. L8N3025-1🕒	0.90	“ angle”“ angles”“ triangles”
4. L6N3528-1🕒	0.88	“ angles”“ angle”“ triangles”
5. L2N4629+1💥	0.88	“ trig”“ Trig”“ angles”
6. L7N1106-1🕒	0.86	“ angle”“ angles”“ Angle”
7. L6N1084-1🕒	0.86	“ angle”“ angles”“ angular”
8. Sohcahtoa✅ L6N1925-1	0.85	“ math”“ triangles”“ angle”
9. L18N4438+1🕒	0.81	“ angle”“ angles”“ Angle”
10. L23N4645-1🕒	0.81	“ angles”“ angle”“ Angle”

Shadow specimen, apart from the ten: L3N1506-1🕒, -0.14.

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 3, neuron 1506, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.73, top-1 agreement 0.31) 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
L3N1506+1 t0-v0.1	3.734	0.31	4.316	2.83	Top boosted tokens ↑
L3N1506-1🕒 shadow specimen t0-v0.1	1.057	0.62	1.253	1.53	“ reference”“⏎”“ ‘”“ code”“ “”“ machine”“ referenced”“ refer”

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.