INDEX

Explanations

collect and categorize information

language and context shifts

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

evenly

0.47

nan

0.46

с

0.45

eta

0.43

between

0.43

reset

0.43

сред

0.43

acion

0.42

equal

0.42

replace

0.41

POSITIVE LOGITS

كلم

0.52

 Thema

0.47

 Künst

0.46

 steckt

0.44

 flaws

0.44

Mé

0.43

 nuov

0.42

 Turismo

0.42

ల్ని

0.42

 Flore

0.41

Activations Density 0.000%