INDEX

Explanations

gradual change or progression

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

↵↵

-1.89

有一個

-1.56

Ő

-1.38

IMUM

-1.37

 hoher

-1.35

 wichtiger

-1.35

 совме

-1.34

 solicitó

-1.33

-1.32

жений

-1.32

POSITIVE LOGITS

蘩

1.88

."

1.76

,”

1.75

</h3>

1.73

袿

1.60

rías

1.59

 slowly

1.55

1.54

лів

1.54

 gradually

1.52

Activations Density 0.023%