INDEX

Explanations

specific items and their components

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

 even

-1.66

 just

-1.50

 because

-1.41

 like

-1.36

ский

-1.30

 már

-1.30

 donnera

-1.27

asnya

-1.27

 لأنه

-1.26

𝐋

-1.24

POSITIVE LOGITS

٨

1.52

Then

1.49

mäß

1.41

ſy

1.41

and

1.35

٧

1.34

 також

1.34

 asus

1.34

 także

1.31

 också

1.30

Activations Density 0.206%