INDEX

Explanations

training courses and programs

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

 ChatGPT

-1.53

 YouTuber

-1.51

 reddit

-1.48

 empêcher

-1.45

https

-1.44

-1.42

 TikTok

-1.42

 Millennials

-1.40

 certamente

-1.39

POSITIVE LOGITS

 their

1.72

 although

1.70

 these

1.69

 mainly

1.48

 particularly

1.45

on

1.43

 they

1.41

 approximately

1.41

 seemed

1.38

 other

1.38

Activations Density 0.026%