INDEX

Explanations

variety and technology review

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

aul

2.67

nown

2.39

នុស្ស

2.30

ו

2.27

реже

2.27

свя

2.23

ніципа

2.20

үл

2.20

⿵

2.20

umption

2.17

POSITIVE LOGITS

da

1.81

se

1.56

da

1.52

som

1.47

bes

1.44

 dint

1.40

 gestern

1.36

 medlem

1.35

doy

1.34

Activations Density 0.001%