INDEX

Explanations

markdown or code blocks

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

ேத்க

2.65

 调用

2.57

 광고

2.56

Getenv

2.55

 ヘッド

2.51

 기업

2.51

<unused1368>

2.47

已

2.46

 コーヒー

2.44

 출시

2.43

POSITIVE LOGITS

amin

1.41

ty

1.38

<strong>

1.38

ations

1.24

<b>

1.24

tent

1.10

ties

1.08

ry

1.06

<bos>

1.04

Activations Density 0.159%