INDEX

Explanations

interpreted as

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

ermo

1.30

 inevitably

1.28

 Evidently

1.27

Sal

1.24

Ing

1.24

 Ruskin

1.21

 стреми

1.19

Hj

1.17

Fal

1.16

 कर्तव्य

1.14

POSITIVE LOGITS

也可以

1.76

也可

1.59

 anywhere

1.54

하거나

1.52

 fácilmente

1.52

easily

1.45

 любого

1.45

if

1.44

 dowol

1.43

 easily

1.42

Activations Density 0.980%