INDEX

Explanations

believe what, yield savings

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

 throttling

1.33

 slashing

1.30

 Ducks

1.20

Lil

1.19

elu

1.18

Circuit

1.18

circ

1.14

 taman

1.11

Continu

1.11

 damned

1.11

POSITIVE LOGITS

 रियलिटी

1.77

据

1.70

 रिए

1.61

页

1.61

சிரியர்

1.60

 Τα

1.58

μφωνα

1.56

 情報

1.53

 ಅವರಿಗೆ

1.51

으면

1.50

Activations Density 0.069%