INDEX

Explanations

geographical locations

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

 reason

0.66

 believing

0.63

 certainty

0.62

 contradicting

0.62

 somewhat

0.62

reason

0.61

words

0.60

 justifying

0.59

attempt

0.58

 conflicting

0.58

POSITIVE LOGITS

 Üniversitesi

0.97

 Avenue

0.93

ische

0.89

ское

0.86

 fundada

0.86

 város

0.85

城的

0.85

 berbasis

0.85

isches

0.85

 लौटे

0.84

Activations Density 0.692%