INDEX

Explanations

specific object parts

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

 utilisez

-1.45

 kteří

-1.40

Tils

-1.38



-1.36

óság

-1.34

 говорил

-1.33

Bety

-1.33

Mvh

-1.31

그러나

-1.30

 parís

-1.30

POSITIVE LOGITS

for

1.52

is

1.48

 itself

1.46

 also

1.32

now

1.28

 другие

1.28

 ahora

1.27

itself

1.26

 sadece

1.23

 только

1.22

Activations Density 0.126%