INDEX

Explanations

and or

and, or, finally, even

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 defects

0.42

 increasing

0.41

 trip

0.41

を除く

0.40

:-)

0.40

።

0.39

दित

0.38

 nhỉ

0.38

ε

0.38

 westward

0.38

POSITIVE LOGITS

 lastly

1.00

 thậm

0.91

 hatta

0.91

 даже

0.85

 Lastly

0.83

 sogar

0.83

 infine

0.80

甚至是

0.80

 এমনকি

0.79

Lastly

0.77

Activations Density 0.168%