INDEX

Explanations

method definition prefixes

the pattern "super." in code, which calls parent class methods.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 siquiera

-1.81

 prácticamente

-1.81

 její

-1.79

ڧ

-1.61

頭に

-1.55

𞤤

-1.52

𞤵

-1.51

 siguiendo

-1.47

܃

-1.46

 econó

-1.45

POSITIVE LOGITS

as

1.95

this

1.88

of

1.84

to

1.79

since

1.78

!”

1.68

it

1.65

1.61

ns

1.60

uk

1.59

Activations Density 0.005%