INDEX

Explanations

positional or sequential indicators after "the"

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

不说

1.24

смотря

1.22

ทุก

1.15

тится

1.14

มาก

1.13

sia

1.12

 وي

1.11

状態

1.10

Ně

1.09

 خپل

1.08

POSITIVE LOGITS

why

1.42

phosphine

1.42

how

1.38

 truncated

1.32

 discoloration

1.31

 emergence

1.27

 expanded

1.26

 oppression

1.26

 redness

1.26

 rocker

1.26

Activations Density 0.437%