INDEX

Explanations

code block structure

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

migrationBuilder

-1.37



-1.30

volle

-1.27

Tetapi

-1.27

 odpově

-1.27

provided

-1.26

 accompagner

-1.25

 metic

-1.23

specifically

-1.23

 impecable

-1.23

POSITIVE LOGITS

by

1.67

 with

1.62

 from

1.55

It

1.44

 always

1.41

 mainly

1.38

 With

1.36

 were

1.34

 without

1.34

1.30

Activations Density 0.033%