INDEX

Explanations

news outlets and specific entities

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

 artículo

-1.57

 cristianos

-1.55

 particuli

-1.55

 také

-1.55

鸶

-1.54

efter

-1.53

囮

-1.50

</sup>

-1.46

 tragé

-1.45

 officier

-1.44

POSITIVE LOGITS

 when

2.28

 what

2.08

how

1.89

 other

1.84

it

1.82

 people

1.61

 both

1.50

 this

1.49

 those

1.49

When

1.48

Activations Density 0.005%