INDEX

Explanations

humanity and self-reflection

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

remarkable

1.46

Despite

1.43

Although

1.39

 मानदंडों

1.32

typical

1.30

 Despite

1.26

Naturally

1.25

Shortly

1.25

southern

1.21

 समृद्ध

1.19

POSITIVE LOGITS

 też

1.38

 pula

1.27

 ALSO

1.26

 आल्सो

1.12

 also

1.10

 likewise

1.09

াও

1.05

 else

1.05

もお

1.04

 others

1.03

Activations Density 4.295%