INDEX

Explanations

providing services, guidance, or resources

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

但是在

-1.73

 when

-1.60

 while

-1.59

因为它

-1.57

 οποία

-1.56

ungguh

-1.45

それは

-1.39

较高

-1.39

if

-1.34

但在

-1.33

POSITIVE LOGITS

3.33

an

2.11

 such

1.56

 enough

1.55

#%%

1.54

 trajets

1.52

 coelho

1.46

 óculos

1.44

 lápis

1.43

 fausse

1.42

Activations Density 0.064%