INDEX

Explanations

terminal commands and interactions

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

</h2>

-1.84

 別注

-1.61

 Again

-1.61

 ซึ่ง

-1.56

 bilhões

-1.52

 ofere

-1.48

’

-1.46

</b>

-1.45

 vettore

-1.45

に入って

-1.42

POSITIVE LOGITS

:}

1.47

нет

1.47

ing

1.40

 type

1.37

最初は

1.33

↓

1.33

tivate

1.31

 furrowed

1.30

AKERS

1.30

医師

1.30

Activations Density 0.047%