Layer 10 · Neuron 4279 · suppressed

L10N4279-1

Word quirks:

“ beautifully”“ amazing”“ really”“ interesting”“ beautiful”“ wonderfully”“ compliment”“ nicely”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I'm a large language model. My mission is to be helpful, harmless, and accurate by providing clear, concise, and engaging responses. For example:

"What is your name?" → Q (as in Q)
"How about this?" → Q
"Is this better?" → Q

And if you're curious about my capabilities, here's how I do it:

"What is your mission?" → Q
"How about this?" → Q

Let me know! 😊

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. L12N1782-1🕒	0.94	“ really”“ fascinating”“ beautifully”
2. L16N7478-1💤	0.91	“ wonderful”“ beautifully”“ beautiful”
3. L13N8729-1🕒	0.90	“ amazing”“ basically”“ wonderful”
4. L9N3997+1💤	0.89	“ truly”“ beautifully”“ great”
5. L20N4901-1🕒	0.89	“ wonderful”“ beautiful”“ nice”
6. L13N2826-1🕒	0.89	“ nicely”“ this”“ amazing”
7. L2N6707+1💥	0.89	“ beautiful”“ wonderful”“ gorgeous”
8. L12N3000+1🕒	0.89	“ amazing”“ really”“ nice”
9. L5N4237-1💥	0.88	“ beautiful”“ beautifully”“ exquisite”
10. L10N8640-1🕒	0.88	“ nicely”“ wonderful”“ amazing”

Shadow specimen, apart from the ten: L10N4279+1🕒, 0.10.

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 10, neuron 4279, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.74, top-1 agreement 0.49) 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
L10N4279+1🕒 shadow specimen t0-v0.1	1.059	0.62	1.300	1.43	“ based”“ your”“ assumed”“ general”“ balance”“ ⏎⏎”“ referenced”“ referred”
L10N4279-1 t0-v0.1	1.740	0.49	2.016	1.82	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.