INDEX

Explanations

emails and contact info

commands and operations

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ق

1.53

ل

1.52

ي

1.50

на

1.49

for

1.45

and

1.41

મ

1.38

و

1.37

ли

1.35

ه

1.31

POSITIVE LOGITS

1.34

to

1.27

↵↵

1.22

1.19

1.13

is

1.10

，

1.10

1.06

1.05

,\\

0.96

Activations Density 0.157%