INDEX

Explanations

used

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

to

-0.78

 nahilalakip

-0.63

 hâte

-0.62

 vendus

-0.62

 medlemmer

-0.58

 larmes

-0.58

 difficultés

-0.58

 chiaramente

-0.56

 lágrimas

-0.56

 puisse

-0.55

POSITIVE LOGITS

 AssemblyTitle

0.86

 &___

0.69

 CreateTagHelper

0.61

 समीक्षाओं

0.59



0.58

IBOutlet

0.57

AccessorTable

0.56

.*")]

0.56

 productive

0.53

++

0.53

Activations Density 0.053%

No Known Activations