INDEX

Explanations

captures what is apparent

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

<unused646>

1.61

 Beyonce

1.57

<unused1910>

1.51

<unused694>

1.48

 Millennials

1.47

 Shit

1.46

 multiverse

1.46

<unused418>

1.45

<unused714>

1.44

<unused457>

1.44

POSITIVE LOGITS

らかじめ

0.91

ży

0.88

ুব

0.87

 geldig

0.87

mann

0.86

ട്ട

0.86

它可以

0.84

achte

0.81

ীর

0.79

Activations Density 0.185%