INDEX

Explanations

evaluating likes and dislikes

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

پ

1.35

ト

1.24

ította

1.22

ペ

1.21

ک

1.21

Thể

1.21

 zawiera

1.21

 Terbaru

1.21

 meliputi

1.20

ウ

1.20

POSITIVE LOGITS

 gradual

1.27

 greasy

1.26

 egalitarian

1.22

 mundane

1.21

 inconven

1.19

 unsightly

1.19

 abrupt

1.18

 belliger

1.17

 unpleasant

1.15

 microscopic

1.15

Activations Density 0.356%