INDEX

Explanations

autoregressive process

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

 זו

1.04

les

0.99

 course

0.92

 otherwise

0.92

נט

0.89

une

0.87

 Otherwise

0.87

 alternatively

0.84

 zone

0.83

 easier

0.82

POSITIVE LOGITS

に至

1.47

finaly

1.28

 culminates

1.25

 sampai

1.25

Finale

1.25

 einschließlich

1.24

finale

1.23

 culminating

1.19

conclusion

1.19

までも

1.19

Activations Density 0.367%