INDEX

Explanations

references to a specific individual named Will

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Did

-0.97

Did

-0.94

Have

-0.68

 Have

-0.61

Has

-0.57

Does

-0.54

 Does

-0.52

Got

-0.52

Has

-0.51

Do

-0.48

POSITIVE LOGITS

 Will

1.90

Will

1.63

WILL

1.48

 WILL

1.15

 Wills

1.01

ScopeManager

0.83

twimg

0.81

 kasarigan

0.79

 Willen

0.79

 wills

0.78

Activations Density 0.002%