INDEX

Explanations

contrasting ideas

contrast or consequence followed by "This"

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

きっと

0.45

ण्यास

0.45

าศ

0.43

้าง

0.43

ascii

0.42

 laatste

0.40

今は

0.40

ಭಾಗ

0.40

次は

0.39

鈾

0.39

POSITIVE LOGITS

 unlike

0.71

不像

0.65

unlike

0.63

 Unlike

0.59

Unlike

0.58

 therefore

0.55

因而

0.55

 compared

0.52

 поэтому

0.52

因此

0.52

Activations Density 0.044%