INDEX

Explanations

upon

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 upon

-1.08

upon

-0.97

 UPON

-0.97

Upon

-0.84

 démocr

-0.83

 among

-0.80

 Upon

-0.79

 turísticos

-0.74

 engraçadas

-0.73

 enfans

-0.71

POSITIVE LOGITS

 experimental

0.71

the

0.71

 different

0.63

 transgenic

0.60

 them

0.58

 Experimental

0.57

 cerebral

0.55

 other

0.55

 final

0.54

 experimentally

0.54

Activations Density 0.330%