INDEX

Explanations

here is

model response preamble

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

 অসু

0.31

 musste

0.31

 দিত

0.30

 adoles

0.30

விய

0.30

침

0.30

 피해

0.30

 abstrak

0.30

경

0.30

 astron

0.30

POSITIVE LOGITS

HERE

0.42

’

0.39

here

0.39

 HERE

0.37

Here

0.37

is

0.35

son

0.34

 Voici

0.33

fordshire

0.32

么

0.32

Activations Density 0.045%