INDEX

Explanations

code-related keywords such as "join" and "switch"

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

原始内容存档于

-0.89

archiviato

-0.84

 للمعارف

-0.77

 AssemblyCulture

-0.76

Datuak

-0.73

Lähteet

-0.72

BASEPATH

-0.71

UserScript

-0.71

 kaynağından

-0.69

DeleteBehavior

-0.69

POSITIVE LOGITS

<bos>

1.12

х

0.46

<tr>

0.45

ヘル

0.45

 expressed

0.44

 ifrån

0.43

yntaxException

0.43

fast

0.41

har

0.41

Activations Density 0.084%