INDEX

Explanations

can speak chinese, src paths

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

臾

2.78

𝑢

2.58

ਲੇ

2.54

из

2.49

 اړه

2.48

.’”

2.47

ه‌

2.46

etting

2.46

eds

2.46

$$\

2.45

POSITIVE LOGITS

 Informe

2.98

ียร์

2.43

 साध

2.42

rans

2.39

 entgegen

2.38

ributed

2.37

pil

2.35

("

2.32

◅

2.32

 Diarsipkan

2.29

Activations Density 0.003%