INDEX

Explanations

actions, processes, and schemes

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

`"

2.73

玾

2.65

*/

2.56

`/

2.53

`#

2.51

\/}

2.47

`./

2.42

`.

2.41

 使っ

2.41

`{

2.40

POSITIVE LOGITS

atomy

2.79

обходимо

2.73

xiety

2.66

ногда

2.41

frast

2.38

спользу

2.37

oltre

2.36

withstanding

2.35

quake

2.25

spapers

2.23

Activations Density 2.144%