INDEX

Explanations

acronyms and specific names

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

 Répondre

-2.39

ኧ

-2.39

 古风

-2.39

 Objetivos

-2.38

 撮っ

-2.38

 Preguntas

-2.34

odeon

-2.31

毟

-2.31

 anún

-2.30

ቼ

-2.30

POSITIVE LOGITS

end

2.94

jaqueta

2.58

of

2.52

</strong>

2.48

al

2.47

2.41

 united

2.33

蜮

2.28

 cherished

2.27

Activations Density 0.056%