INDEX

Explanations

data structures and sequences

The neuron is highly activated by floating‐point numbers (decimal numerals) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 İng

0.88

 Oyun

0.80

 Emmy

0.75

 parco

0.73

Địa

0.72

 compartió

0.72

珏

0.71

ের

0.70

 comparten

0.70

 gồm

0.70

POSITIVE LOGITS

ка

0.84

К

0.83

 став

0.80

Т

0.80

오늘

0.78

т

0.78

..........

0.77

 acts

0.75

 поиск

0.75

Т

0.74

Activations Density 0.000%