INDEX

Explanations

code structure and configuration

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

 huống

1.35

लेशन

1.27

 základě

1.25

ơn

1.22

 zusätz

1.22

 nhiệm

1.19

েশনে

1.19

áver

1.15

ávan

1.14

ဆံ

1.14

POSITIVE LOGITS

1.70

،

1.64

、

1.43

 galore

1.31

，

1.23

tog

1.22

,...,

1.21

 constitute

1.12

،

1.10

 onboard

1.10

Activations Density 0.000%