INDEX

Explanations

the word "chip"

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TBC

-0.79

Życiorys

-0.79

 viscose

-0.78

 Lining

-0.68

Enllaços

-0.66

 Loreto

-0.65

Jah

-0.64

TBC

-0.64

 Enclosure

-0.63

 ligiloj

-0.63

POSITIVE LOGITS

 chips

1.93

 Chips

1.81

 Chip

1.78

 chip

1.76

Chips

1.73

chips

1.61

Chip

1.52

chip

1.50

 CHIP

1.46

 chipped

1.31

Activations Density 0.006%