INDEX

Explanations

grammatical corrections and explanations

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

 감소

1.36

1.29

Rodr

1.29

IZATION

1.28

 보호

1.27

ល្អ

1.26

शेखर

1.26

 설명

1.25

ស្រាយ

1.24

Abdul

1.22

POSITIVE LOGITS

1.22

 certa

1.03

 furt

0.99

 quelque

0.95

ou

0.94

 parte

0.90

to

0.87

out

0.85

the

0.84

at

0.83

Activations Density 0.082%