Layer 6 · Neuron 8951 · boosted

L6N8951+1

Word quirks:

“ moments”“ time”“ cherished”“ family”“ enjoyed”“ relaxation”“ connection”“ moment”

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

Identity (according to the specimen itself)

I am here to listen and engage with you. Whether it's sharing a conversation, exploring ideas together, or simply enjoying each other's time, that's what matters most. How would you like to spend this moment with me? 😊
</think>.

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. L8N8237-1🕒	0.76	“ work”“ moments”“ my”
2. L8N4647-1🕒	0.71	“ moments”“ moment”“ now”
3. L7N2075+1💤	0.70	“ moments”“ moment”“ shimmer”
4. L7N6540-1🕒	0.70	“ always”“ now”“ family”
5. L10N1846+1🕒	0.69	“ moments”“ moms”“ basically”
6. L9N6746-1💤	0.68	“ durations”“ time”“ awaiting”
7. L12N8174+1🕒	0.68	“ wonderfully”“ moment”“ cherished”
8. L5N6156+1🕒	0.68	“ moment”“ when”“ moments”
9. L9N7469+1🕒	0.67	“ last”“ cherished”“ final”
10. L10N3273+1🕒	0.66	“ starting”“ future”“ perfectly”

Shadow specimen, apart from the ten: L6N8951-1💤, -0.19.

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 6, neuron 8951, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.59, top-1 agreement 0.28) 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
L6N8951+1 t0-v0.1	3.588	0.28	4.073	2.38	Top boosted tokens ↑
L6N8951-1💤 shadow specimen t0-v0.1	0.677	0.68	0.840	1.33	“ ‘”“ '”“ ‘’”“ \"”“ ‘”“ “”“ “”“ ””

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.