INDEX

Explanations

not when

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

−</

1.56

రుణ

1.54

GMT

1.52

Cors

1.45

Cort

1.42

Tart

1.37

 прав

1.36

Trails

1.33

Prairie

1.33

))))

1.32

POSITIVE LOGITS

 aluminium

1.27

 estrella

1.19

 codigo

1.17

 aluminio

1.17

骁

1.16

 anteriormente

1.15

 ceramic

1.15

 rapeseed

1.14

 coinChange

1.13

сунок

1.11

Activations Density 0.001%