INDEX

Explanations

True or False

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

 subpo

1.39

 annak

1.27

 subpoena

1.26

 langs

1.23

 aliran

1.23

 Provinsi

1.21

请求

1.20

zg

1.19

 telepon

1.17

क्सी

1.16

POSITIVE LOGITS

 millennia

1.58

 beamed

1.52

 تصميم

1.49

 μεγάλη

1.48

карбо

1.47

Whit

1.46

optimized

1.44

fuzzy

1.44

feather

1.42

ისტ

1.41

Activations Density 0.063%