INDEX

Explanations

copyright holders and proper nouns

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

 intér

-1.55

物体

-1.41

IBinder

-1.40

corrhi

-1.39

izier

-1.38

᎒

-1.37

<",

-1.35

 Homem

-1.33

 száll

-1.30

 sabor

-1.30

POSITIVE LOGITS

휀

1.68

与此同时

1.66

ですが

1.61

があり

1.57

<bos>

1.56

 reveló

1.44

 weihnachten

1.41

更何况

1.40

偶尔

1.40

to

1.39

Activations Density 0.000%