INDEX

Explanations

often followed by descriptive adjectives

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

 astronomical

1.60

及ひ

1.59

 Ще

1.59

טה

1.59

𝑐

1.58

ycin

1.57

𝙖

1.55

yama

1.54

𝙞

1.54

𝙨

1.52

POSITIVE LOGITS

 vezető

1.47

 garantiert

1.41

argar

1.41

furan

1.40

 সমিতির

1.39

ক্তার

1.35

첨

1.35

uploaded

1.35

ধর্ম

1.35

木

1.34

Activations Density 0.090%