INDEX

Explanations

positive evaluations

good for starting

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 صحیح

0.38

 oldukça

0.38

anone

0.36

 "").

0.34

 apparaissent

0.33

 প্রকার

0.33

 অথবা

0.31

 quelconque

0.31

 काफी

0.31

으로는

0.31

POSITIVE LOGITS

 idea

0.66

 little

0.61

little

0.60

 example

0.59

ulously

0.59

 memories

0.56

 craftsmanship

0.56

 opportunity

0.56

 accomplishment

0.55

idea

0.54

Activations Density 0.612%