INDEX

Explanations

references to the word "As" used to introduce comparisons or conditional statements

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

DateTimeField

-0.84

RenderAtEndOf

-0.84

LikeLike

-0.82

MLLoader

-0.78

 houſe

-0.76

 Houſe

-0.76

DebuggerNonUser

-0.76

Ự

-0.75

 ſtre

-0.73

venidos

-0.73

POSITIVE LOGITS

as

2.39

As

1.88

As

1.74

AS

1.63

as

1.46

 ως

1.23

 sebagai

1.16

 как

1.16

 як

1.13

作為

1.10

Activations Density 6.005%