INDEX

Explanations

once followed by a state change

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

on

-1.60

 that

-1.59

 been

-1.46

 while

-1.37

 before

-1.34

 since

-1.33

 come

-1.31

 einfachen

-1.30

as

-1.29

⾦

-1.26

POSITIVE LOGITS

初め

1.33

ärm

1.32

 nimmer

1.31

 konfig

1.30

 يقول

1.29

２

1.23

läs

1.22

instead

1.21

culo

1.21

glycemia

1.20

Activations Density 0.012%