INDEX

Explanations

object-oriented, packet-based, block-based

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sujeito

-0.93

部の

-0.92

 bezeichnen

-0.88

 футбо

-0.87

vaient

-0.86

onomie

-0.83

interopRequire

-0.83

 informacija

-0.82

 compartió

-0.81

ঔ

-0.81

POSITIVE LOGITS

 oriented

1.38

 based

1.24

 orient

1.23

wise

1.21

 Based

1.18

 Oriented

1.10

Based

1.09

oriented

1.09

 wise

1.07

Oriented

1.02

Activations Density 0.163%