INDEX

Explanations

Okay, let's break down

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 quarries

0.57

 ग्

0.55

 plotting

0.55

 renting

0.55

 playerId

0.54

 quarry

0.54

cue

0.54

 park

0.53

 plaintiff

0.53

 fringe

0.52

POSITIVE LOGITS

dihydro

0.66

 টেস্ট

0.63

TEST

0.61

Certainly

0.60

 uzyska

0.59

gener

0.59

 Certainly

0.57

immunity

0.57

有一個

0.57

Türk

0.57

Activations Density 0.000%