INDEX

Explanations

when describing states

describing states or conditions

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

 некоторых

0.50

 някои

0.47

 illetve

0.46

OF

0.46

某些

0.46

 niektórych

0.44

 সংক্রান্ত

0.43

moe

0.42

 varies

0.42

 многих

0.40

POSITIVE LOGITS

 असताना

0.69

 ketika

0.59

だったので

0.57

 cuando

0.55

当时

0.55

 när

0.54

 damals

0.52

 quando

0.52

 wollte

0.52

でしたが

0.52

Activations Density 0.004%