INDEX

Explanations

This neuron does not activate on any input—it’s effectively “dead” and isn’t detecting any pattern.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

۷

0.83

৭

0.82

 সাড়ে

0.80

९

0.77

 साढ़े

0.75

৬

0.71

७

0.71

７

0.71

９

0.69

۶

0.66

POSITIVE LOGITS

 tiny

1.12

 minuscule

0.98

tiny

0.95

小さな

0.91

 kecil

0.89

 small

0.87

 Tiny

0.86

 néhány

0.86

 pequena

0.86

 infinitesimal

0.86

Activations Density 0.615%