INDEX

Explanations

file names or code structures

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

ploidy

1.31

christmas

1.28

óc

1.22

在北京

1.22

cpf

1.20

在了

1.19

εις

1.18

在你

1.17

在我们

1.17

 στον

1.16

POSITIVE LOGITS

 וב

1.68

 midst

1.48

 regard

1.46

 spite

1.43

 outburst

1.42

 условиях

1.42

 surroundings

1.37

 режиме

1.30

 conjunction

1.30

 البلاد

1.30

Activations Density 2.180%