INDEX

Explanations

instruction and technical specifications

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Teach

-0.84

 Teach

-0.79

oredCriteria

-0.70

 ―――――

-0.68

 GenerationType

-0.68

}))

-0.66

RTDA

-0.66

parsedMessage

-0.65

Coat

-0.64

UseVisualStyle

-0.60

POSITIVE LOGITS

Me

0.51

 Bíblia

0.45

ME

0.44

the

0.44

+'/

0.44

RegressionTest

0.43

ellation

0.43

Me

0.42

field

0.42

 '+':

0.42

Activations Density 0.887%