INDEX

Explanations

jupyter, python, notebook, colab

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-1.88

-1.80

 from

-1.71

甃

-1.67

 Their

-1.63

盌

-1.63

 kurie

-1.59

$,

-1.58

瑨

-1.58

 There

-1.54

POSITIVE LOGITS

 что

1.59

男の子

1.53

ではまた

1.43

明日は

1.42

 mangled

1.41

わたしは

1.40

६

1.40

スポンサーリンク

1.39

荑

1.38

死ぬ

1.36

Activations Density 0.029%