INDEX

Explanations

as needed

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

if

1.33

really

1.31

 yourself

1.26

 really

1.23

 anything

1.18

 ANYTHING

1.15

anything

1.15

}$-

1.12

 yourselves

1.12

yourself

1.11

POSITIVE LOGITS

 часть

2.29

 Teile

2.27

 parte

2.16

部分

2.08

 część

2.04

 رابطه

2.00

PartResult

1.98

 текст

1.93

 part

1.93

 نسبة

1.90

Activations Density 0.130%