INDEX

Explanations

importance and certainty about what follows

verbs followed by "to"

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

 ইত্যা

0.33

 nhưng

0.33

 пенсии

0.31

 বিয়ের

0.30

 இருந்தாலும்

0.29

ఓ

0.29

 यात

0.29

 utilisent

0.29

 этом

0.29

 घात

0.29

POSITIVE LOGITS

to

0.55

 να

0.51

 upang

0.46

to

0.43

 bahwa

0.40

ที่จะ

0.39

 that

0.38

 kuti

0.37

that

0.37

 rằng

0.37

Activations Density 0.306%