INDEX

Explanations

human doesliterally translates

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

Brows

1.57

CDATA

1.57

sil

1.56

sag

1.47

 healed

1.44

AnimationFrame

1.43

 denials

1.39

ยก

1.39

ชอบ

1.37

 caut

1.36

POSITIVE LOGITS

pples

2.05

 οποίο

2.05

 ਪ੍ਰ

1.91

ጠት

1.86

omyces

1.84

名

1.76

 afterPath

1.72

atrième

1.69

ང

1.68

ＰＥ

1.67

Activations Density 0.051%