INDEX

Explanations

percentages or counts

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

痈

-1.35

ラックス

-1.32

鶫

-1.28

 sitten

-1.26

ßten

-1.23

 interesses

-1.22

mittelt

-1.22

 Però

-1.21

ρέπει

-1.20

荸

-1.20

POSITIVE LOGITS

one

2.13

 only

1.79

1.63

two

1.60

 first

1.54

1.45

few

1.30

 small

1.30

or

1.25

 comenz

1.23

Activations Density 0.094%