INDEX

Explanations

numbers followed by i

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

േറ്റ

0.91

 Saviour

0.91

рому

0.89

 ያላቸው

0.87

შე

0.86

entra

0.86

ringle

0.86

라면

0.86

врат

0.86

 Chop

0.85

POSITIVE LOGITS

THAT

1.39

 that

1.38

 THAT

1.30

 rằng

1.24

 WHICH

1.24

that

1.24

 which

1.23

That

1.16

que

1.16

 That

1.09

Activations Density 0.434%