INDEX

Explanations

question and answer markers

lists of items

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

↵↵

0.82

↵↵↵

0.66

↵↵↵↵

0.62

<start_of_image>

0.49

↵↵↵↵↵

0.46

 }}$.

0.46

↵↵↵↵↵↵↵

0.43

امر

0.42

 systemic

0.41

鷺

0.41

POSITIVE LOGITS

 यही

0.46

":[],

0.42

 নির্মাতা

0.41

 అవి

0.40

keydown

0.39

 убе

0.38

kur

0.37

दारों

0.37

 McCabe

0.36

assertInstanceOf

0.36

Activations Density 0.001%