INDEX

Explanations

number followed by one or ten

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 الثاني

1.44

 دوسرا

1.40

 Largest

1.39

Second

1.34

 Ones

1.33

 Pertama

1.32

 Shaped

1.32

 deuxième

1.31

ראש

1.29

 Smaller

1.28

POSITIVE LOGITS

Fel

1.61

 Matth

1.47

Fel

1.46

memo

1.46

Gom

1.45

ped

1.43

fel

1.42

Bern

1.41

 beet

1.39

fel

1.39

Activations Density 0.004%