INDEX

Explanations

closing punctuation and code constructs

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

บรร

-0.94

 their

-0.91

 tarvit

-0.85

 viera

-0.85

 fotos

-0.82

 feminina

-0.82

 paulista

-0.82

 même

-0.80

чный

-0.79

 jardin

-0.79

POSITIVE LOGITS

這就是

0.89

ffold

0.89

κο

0.85

 resourceCulture

0.84

 古い

0.83

そんな

0.83

 новую

0.83

);

0.82

)։

0.82

=>

0.82

Activations Density 0.001%