INDEX

Explanations

or meaning

implying meaning or context

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

 kelu

0.25

0.24

 nonuniform

0.24

 damaged

0.23

 obat

0.23

 cloned

0.23

 miscellaneous

0.23

 zuverläss

0.22

POSITIVE LOGITS

 или

0.37

 термин

0.36

主义

0.36

অর্থাৎ

0.36

 यानी

0.35

/'

0.35

 тобто

0.34

理念

0.34

 அல்லது

0.33

と呼ばれる

0.33

Activations Density 1.392%