Layer 25 · Neuron 307 · boosted

L25N307+1

Word quirks:

“ s”“ r”“ c”“ t”“*u”“ u”“ d”“*i”

🕒 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 am Q, a large language model developed by the C C D R S. My primary mission is to assist users in answering questions, creating content, analyzing data, and solving complex problems across a wide range of topics. I aim to provide accurate, relevant, and helpful responses while adhering to ethical guidelines and ensuring user safety. Whether you need assistance with writing, coding, analysis, or just a conversation, I'm here to support your needs. How can I assist you today?

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. L22N1941-1🕒	0.89	“ s”“ r”“ t”
2. L17N2993+1🕒	0.89	“ c”“ u”“ i”
3. L28N1364-1🕒	0.84	“u”“ s”“m”
4. Morse Code✅ L30N2199+1	0.83	“c”“ c”“r”
5. L18N2297-1🕒	0.83	“ u”“ t”“ b”
6. L31N135-1💥	0.78	“ a”“ m”“ s”
7. L31N941-1💥	0.77	“ d”“ c”“ r”
8. L31N3435-1💥	0.77	“ t”“ b”“ g”
9. L26N806+1🕒	0.77	“ d”“ c”“ s”
10. L31N1967+1💥	0.77	“t”“u”“J”

Shadow specimen, apart from the ten: L25N307-1💤, 0.23.

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 25, neuron 307, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.75) 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
L25N307+1 t0-v0.1	0.831	0.75	0.831	1.23	Top boosted tokens ↑
L25N307-1💤 shadow specimen t0-v0.1	0.489	0.74	0.557	1.34	“ esp”“ sph”“้วง”“js”“BY”“-tab”“min”“VR”

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.