INDEX

Explanations

acronyms followed by punctuation

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

 étrangère

-1.52

 efficaces

-1.38

 faciles

-1.35

</b>

-1.31

danke

-1.29

-1.27

 échou

-1.27

つづく

-1.26

実際の

-1.23

Какво

-1.23

POSITIVE LOGITS

or

1.69

the

1.52

हरु

1.38

 from

1.36

遽

1.33

或是

1.30

指标

1.28

and

1.26

ógica

1.26

ôtel

1.25

Activations Density 0.176%