INDEX

Explanations

not simple or rarely just

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

chyné

1.67

 particolarmente

1.64

 khususnya

1.58

साइटी

1.58

 lieber

1.53

avits

1.53

 velmi

1.49

uenti

1.48

 misalnya

1.47

连忙

1.45

POSITIVE LOGITS

 ίδια

1.60

 весь

1.57

 ίδ

1.57

 אותו

1.57

<blockquote>

1.53

 aucune

1.52

 nothing

1.46

 spoilt

1.45

 misma

1.43

Absent

1.43

Activations Density 0.322%