INDEX

Explanations

)

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Enllaces

-0.79

<bos>

-0.78

 énergé

-0.77

 définiti

-0.77

ⓧ

-0.75

 sauvages

-0.73

 vectorielles

-0.71

Personensuche

-0.71

 quelcon

-0.71

ніципалі

-0.70

POSITIVE LOGITS

 []).

0.39

0.38

 innov

0.35

 tokenizer

0.34

qrstuvwxyz

0.34

ynchron

0.33

 optimised

0.33

ruptedException

0.33

 boto

0.33

 mais

0.32

Activations Density 0.052%