INDEX

Explanations

potential or possible

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

 insan

1.98

家的

1.73

نا

1.71

،

1.63

рито

1.61

 alas

1.59

let

1.56

ตรี

1.55

ྱ

1.51

 tapp

1.50

POSITIVE LOGITS

ര്‍ട്ട്

2.06

ুষ

1.98

cje

1.95

ྜ

1.87

 geslacht

1.85

 કંઈ

1.85

mce

1.82

luents

1.79

nent

1.79

RELATES

1.77

Activations Density 0.135%