INDEX

Explanations

references to editing or modifications

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

edit

-1.04

Edit

-1.03

 EDIT

-0.96

EDIT

-0.95

 Edit

-0.79

出版年

-0.71

reportWebVitals

-0.70

appears

-0.65

regate

-0.63

abin

-0.61

POSITIVE LOGITS

Editing

2.20

 Editing

1.89

 editing

1.73

editing

1.58

 writing

0.72

Writing

0.63

 WRITING

0.63

writing

0.62

 Writing

0.59

ContentAsync

0.58

Activations Density 0.001%