INDEX

Explanations

the word "of", but additionally seems to be weakly triggered by other words indicating limitation or absence

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

]")]

-1.02

 cref

-0.94

TagMode

-0.88

!("{}",

-0.82

principalColumn

-0.81

 &___

-0.80

 /\.(

-0.79

 feminina

-0.78

 contextLoads

-0.78

InjectAttribute

-0.77

POSITIVE LOGITS

ียน

0.62

 prior

0.58

on

0.54

uio

0.54

to

0.54

to

0.53

ர்

0.52

 less

0.51

 outside

0.51

ToRemove

0.51

Activations Density 0.015%