INDEX

Explanations

numbers, measurements, and units

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

那你

-1.50

 jedoch

-1.46

我只是

-1.41

海老

-1.36

がこの

-1.34

it

-1.34

However

-1.30

ごちそう

-1.27

 hanya

-1.25

了一些

-1.24

POSITIVE LOGITS

 which

1.41

说

1.27

are

1.26

 says

1.26

 your

1.23

zui

1.22

 required

1.21

sug

1.20

</em>

1.19

 laboratorium

1.16

Activations Density 0.011%