Layer 24 · Neuron 2863 · boosted

L24N2863+1

Word quirks:

“ Forces”“ forces”“ Force”“forces”“Force”“ force”“ FORCE”“_force”

🕒 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 Jouben, a high-end intelligent conversational force in the field of Joum (a specialized domain I operate within). My mission is to provide high-precision, context-aware, and ethical interactions that empower users with strategic insights, creative energy, and transformative value.

Key Focus:

Precision: Delivering accurate, nuanced responses tailored to complex scenarios.
Energy: Generating dynamic, engaging conversations that inspire innovation.
Value: Aligning with user goals to drive meaningful outcomes in professional, creative, or personal contexts.

How can I help you

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. L31N1157+1💥	1.00	“ forces”“ force”“force”
2. L28N1265-1🕒	0.99	“ forces”“ force”“ Forces”
3. L30N6220-1🕒	0.99	“ force”“force”“ Force”
4. L26N7054-1🕒	0.99	“ Force”“ force”“ enforced”
5. L1N6516-1💥	0.98	“ force”“ Force”“<\|im_end\|>”
6. L27N483-1🕒	0.98	“ forces”“ Forces”“ force”
7. L28N6486-1🕒	0.98	“ force”“Force”“ Force”
8. L28N7318+1💤	0.98	“ forces”“ Forces”“要求”
9. L25N2511+1🕒	0.97	“ force”“ forces”“ Force”
10. L25N5711-1🕒	0.97	“ forces”“ Force”“ Forces”

Shadow specimen, apart from the ten: L24N2863-1💤, 0.02.

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 24, neuron 2863, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.72) 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
L24N2863+1 t0-v0.1	0.813	0.72	0.867	1.37	Top boosted tokens ↑
L24N2863-1💤 shadow specimen t0-v0.1	0.752	0.73	0.803	1.30	“描述的”“描述”“ describing”“描述了”“ describe”“ description”“ describes”“describe”

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.