INDEX

Explanations

capacity and self

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

 self

-0.53

self

-0.52

bids

-0.50

ModelAdmin

-0.50

ModelSerializer

-0.49

FieldDescriptor

-0.46

 назначения

-0.45

errHandler

-0.45

пусти

-0.45

 Capacity

-0.44

POSITIVE LOGITS

 informée

0.61

IsContent

0.60

 sucrose

0.59

 sphinct

0.59

 estekak

0.58

/*

0.57

 measles

0.57

 chert

0.57

 ویکی‌پدیا

0.57

 برانيه

0.57

Activations Density 0.066%