INDEX

Explanations

all of humanity, members

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

얘

1.56

 معظم

1.54

ったり

1.52

 gitu

1.49

 misalkan

1.47

 مثلا

1.45

쩜

1.42

 تجي

1.39

antra

1.38

 Tycho

1.38

POSITIVE LOGITS

 members

2.51

who

2.45

Members

2.37

 Members

2.25

 participants

2.21

members

2.18

 participating

2.13

 member

2.11

 involved

2.11

who

2.09

Activations Density 0.103%