INDEX

Explanations

Chinese and Japanese particles

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

</h3>

-3.45

</strong>

-2.95

 rivales

-2.63

܆

-2.61

CHREIB

-2.59

む

-2.53

io

-2.52

 peuvent

-2.48

,'

-2.45

новь

-2.42

POSITIVE LOGITS

 This

2.78

tedì

2.50

2.44

is

2.34

 після

2.14

Two

2.13

in

2.09

AutoModerator

2.09

one

2.08

 after

2.08

Activations Density 0.000%