INDEX

Explanations

concepts joined by and

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

 both

-1.95

</strong>

-1.84

насе

-1.61

 such

-1.61

 BOTH

-1.57

 września

-1.55

 сентября

-1.51

licópter

-1.48

dicht

-1.45

 Também

-1.45

POSITIVE LOGITS

of

3.48

1.89

you

1.83

1.74

if

1.71

The

1.65

 that

1.56

↵↵↵↵↵

1.55

Activations Density 0.157%