INDEX

Explanations

code block starters

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

as

-1.51

 will

-1.50

by

-1.43

 would

-1.30

 after

-1.24

 while

-1.23

or

-1.22

 many

-1.20

 from

-1.20

 through

-1.16

POSITIVE LOGITS

apparently

1.27

somebody

1.19

indicate

1.18

seemed

1.18

 muestre

1.18

seems

1.17

weren

1.16

theres

1.15

somewhere

1.11

proved

1.10

Activations Density 0.118%