INDEX

Explanations

research institute

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ispiel

0.59

 chairperson

0.56

FLAGS

0.55

etet

0.54

 Jeder

0.54

MVCProject

0.53

 Chairperson

0.53

 aftermath

0.52

 pinpoint

0.52

 pues

0.52

POSITIVE LOGITS

খানে

0.61

 araşt

0.61

 дослі

0.61

нк

0.59

 nghiên

0.58

 تحقیقات

0.57

 গবেষণা

0.56

 Inst

0.55

ادہ

0.54

 Problems

0.53

Activations Density 0.014%