Layer 26 · Neuron 4098 · boosted

L26N4098+1

Word quirks:

“ routine”“ grou”“ Routine”“ routines”“ Group”“Routine”“ rutina”“routine”

🕒 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 GRO (GuanRongO), a large-scale language model developed by GuanRong Group. My core mission is to solve real-world problems by providing accur, efficient, and safe solutions across diverse scenarios. I aim to:

Enable Grou - Powering the group's operations with intelligent tools and insights.
Solve Problems - Tackle complex challenges in business, science, and daily life.
Create Value - Drive innovation and efficiency for both the group and its users.
Ensure Safety -

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. L30N4780+1💥	0.94	“ routine”“ Routine”“Routine”
2. L24N8209-1🕒	0.93	“ routine”“ Routine”“Routine”
3. L2N8059+1💥	0.90	“ routine”“<\|im_end\|>”“ Routine”
4. L27N3734+1🕒	0.87	“ Routine”“ routine”“Routine”
5. L25N4190-1🕒	0.87	“ routine”“ Routine”“Routine”
6. L19N8781-1💤	0.86	“ routine”“ Routine”“routine”
7. L23N2818-1🕒	0.84	“ routine”“ Routine”“ routines”
8. L20N6403-1💤	0.79	“ routine”“ stages”“ Routine”
9. L29N3681+1💤	0.79	“ habit”“盘”“ routine”
10. L1N6078+1💥	0.78	“<\|im_end\|>”“ routine”“ Routine”

Shadow specimen, apart from the ten: L26N4098-1🕒, -0.03.

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 26, neuron 4098, 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 0.71) 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
L26N4098+1 t0-v0.1	0.813	0.71	0.796	1.35	Top boosted tokens ↑
L26N4098-1🕒 shadow specimen t0-v0.1	0.668	0.71	0.712	1.72	“柄”“意志”“ับ”“总支”“w”“/pkg”“ув”“wand”

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.