INDEX

Explanations

on followed by descriptive phrases

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

 COMMENTS

1.67

침

1.58

문에

1.53

üz

1.48

︎

1.44

Giải

1.41

깔

1.41

 Lied

1.40

 }]);

1.40

πάν

1.39

POSITIVE LOGITS

 behalf

2.76

 occasion

2.04

slaught

2.03

eday

1.94

etime

1.83

 occasions

1.75

boarding

1.73

 tenter

1.71

goin

1.69

omat

1.64

Activations Density 1.597%