INDEX

Explanations

exponents like ^2

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

ЗИ

2.38

<unused826>

2.24

<unused1342>

2.23

<unused1505>

2.22

<unused4>

2.21

<unused1325>

2.19

<unused1339>

2.19

<unused1512>

2.19

<unused1333>

2.18

 pubblico

2.17

POSITIVE LOGITS

ும

1.91

cccccc

1.71

1.68

sez

1.67

DISABLED

1.67

hline

1.65

 일어나

1.64

짖

1.63

<bos>

1.62

Activations Density 0.092%