INDEX

Explanations

widely used or considered

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

ルス

2.18

ructuring

2.12

が無い

1.92

inars

1.92

ahul

1.91

dbc

1.90

hens

1.89

 nanti

1.88

iktar

1.87

始まる

1.86

POSITIVE LOGITS

 regarded

1.58

 considered

1.52

 used

1.47

 revered

1.44

 admired

1.44

regarded

1.43

 widely

1.43

 believed

1.38

Known

1.37

 praised

1.37

Activations Density 0.205%