INDEX

Explanations

the letter 'n' in code

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

 ComVisible

-0.65

Cordialement

-0.65

rouvez

-0.65

()")

-0.63

]){

-0.62

 referenties

-0.62

ImageBackground

-0.61

thyst

-0.60

ysław

-0.60

 }}$}

-0.60

POSITIVE LOGITS

 noms

0.64

0.61

asi

0.61

 Flavor

0.61

 Kasper

0.61

 تضيفلها

0.60

gulier

0.60

lon

0.60

 alias

0.58

 chill

0.57

Activations Density 0.024%