INDEX

Explanations

adored products and solutions

tools, platform, university, challenge

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

 guise

0.39

出来了

0.38

 বিশেষভাবে

0.35

 previamente

0.34

ราม

0.34

 использует

0.34

rapes

0.34

 disguised

0.34

 tạm

0.34

বস

0.34

POSITIVE LOGITS

 adored

1.83

 loved

1.80

Loved

1.65

loved

1.64

 Loved

1.61

 beloved

1.55

 admired

1.54

 frequented

1.48

 enjoyed

1.39

 revered

1.36

Activations Density 0.170%