INDEX

Explanations

the word "Tile"

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

ECL

-0.92

+#+#

-0.91

MCF

-0.89

CloseOperation

-0.82

toBeDefined

-0.81

KBS

-0.80

StructEnd

-0.79

 ujednoznacz

-0.79

ResumeLayout

-0.77

 mohair

-0.77

POSITIVE LOGITS

 Tile

1.48

 tiles

1.42

Tile

1.41

 tile

1.38

tile

1.30

 Tiles

1.22

TILE

1.18

Tiles

1.16

tiles

1.13

 TILE

1.09

Activations Density 0.003%