INDEX

Explanations

numbers and list markers

The neuron strongly activates on numeric tokens (digits and numbers), i.e. it detects occurrences of numerals in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

aqui

-0.97

Aqui

-0.88

 използва

-0.83

 každý

-0.83

党员

-0.82

 damskie

-0.81

OKAY

-0.79

関連項目

-0.78

 Bete

-0.78

 benefícios

-0.78

POSITIVE LOGITS

by

1.34

 द्वारा

1.13

 pelo

0.98

byl

0.96

[];

0.91

 profesora

0.90

enst

0.87

 noviembre

0.87

BY

0.86

 bởi

0.85

Activations Density 0.014%