INDEX

Explanations

keywords related to types and modes

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

-1.63

 that

-1.63

 when

-1.52

and

-1.38

 what

-1.34

on

-1.30

!!

-1.27

!!!!!

-1.23

in

-1.23

 even

-1.23

POSITIVE LOGITS

 спросил

1.42

 immen

1.25

discussed

1.22

 NONE

1.20

nearly

1.20

 HYDRO

1.18

 fervent

1.17

 THROUGH

1.16

 GoPro

1.16

mainly

1.16

Activations Density 0.003%