INDEX

Explanations

dictionary iteration and access

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

しますが

-1.53

-1.52

یں

-1.52

ixante

-1.51

>")

-1.47

--

-1.46

>",

-1.46

;"

-1.44

 dacă

-1.43

toprule

-1.42

POSITIVE LOGITS

of

1.82

 These

1.44

1.39

 medan

1.36

1.34

 diesen

1.33

 цього

1.31

1.29

ze

1.29

 diese

1.26

Activations Density 0.002%