Layer 21 · Neuron 2290 · boosted

L21N2290+1

Word quirks:

“ recall”“recall”“ Recall”“Rec”“ recalled”“回忆”“ recalling”“ remember”

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

Identity (according to the specimen itself)

The reply didn't hold together

Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:
Rec recall:

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. L23N2788+1🕒	0.99	“ recall”“ Recall”“ recalled”
2. L25N5318+1🕒	0.97	“ recall”“ Recall”“Rec”
3. L14N8366-1🕒	0.97	“ recall”“ recalling”“ recalled”
4. L27N2575-1🕒	0.96	“ recall”“ call”“recall”
5. L20N9044-1🕒	0.95	“回忆”“ recall”“ recalling”
6. L9N6369-1🕒	0.95	“ remembered”“ memories”“ remembering”
7. L8N9073-1🕒	0.94	“ recalling”“ recall”“ memory”
8. L19N4773+1🕒	0.94	“ remembered”“ recall”“还原”
9. L9N7185-1🕒	0.94	“ remember”“ remembered”“ remembering”
10. L12N6958+1🕒	0.93	“ recalling”“ recall”“回忆”

Shadow specimen, apart from the ten: L21N2290-1💤, 0.04.

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 21, neuron 2290, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 5.72, top-1 agreement 0.18) 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
L21N2290+1 t0-v0.1	5.717	0.18	6.082	1.31	Top boosted tokens ↑
L21N2290-1💤 shadow specimen t0-v0.1	0.644	0.72	0.762	1.14	“详细描述”“下一条”“向下”“Descriptors”“ description”“ Nar”“azzo”“描述”

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.