INDEX

Explanations

function/filler words

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

 CreateTagHelper

-0.70

 CommonModule

-0.65

(\<

-0.65

^(@)

-0.65

$.

-0.64

.}~\

-0.63

attutto

-0.62

]^{-

-0.62

 BoxFit

-0.61

principalColumn

-0.60

POSITIVE LOGITS

0.81

we

0.65

this

0.57

 this

0.57

it

0.55

we

0.52

それで

0.52

he

0.51

so

0.51

it

0.51

Activations Density 0.064%