INDEX

Explanations

this observation or image

that growth, this makes

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

 الذين

0.14

 Following

0.14

这就是

0.14

 причины

0.13

̓

0.13

💻

0.13

 Introduced

0.13

 sorprend

0.13

 WEEK

0.13

‼

0.13

POSITIVE LOGITS

 seemingly

0.23

 latter

0.23

 newfound

0.23

 tenuous

0.22

 particular

0.21

 burgeoning

0.21

 process

0.20

 aspect

0.20

 precarious

0.20

 disparate

0.20

Activations Density 0.815%