INDEX

Explanations

objects and descriptions

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

すっ

-1.70

マジで

-1.57

垰

-1.48

こいつ

-1.46

 môžete

-1.45

侔

-1.45

 délais

-1.45

ほんとに

-1.42

ごめん

-1.41

鵙

-1.39

POSITIVE LOGITS

 with

2.78

 from

1.53

at

1.52

 eight

1.34

 Eight

1.31

on

1.24

 seven

1.23

 ویتامین

1.22

 appears

1.22

and

1.22

Activations Density 0.003%