INDEX

Explanations

how something will fare or affect

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

成为

-0.88

得到

-0.86

Оста

-0.85

origen

-0.85

才有

-0.85

 ہو

-0.85

 Entstehung

-0.82

हरा

-0.81

 tạo

-0.81

 mít

-0.80

POSITIVE LOGITS

 fared

1.43

 fare

1.38

 react

1.27

 compares

1.23

 compare

1.16

 affect

1.13

 reaccion

1.12

 feel

1.11

 reacted

1.11

 fares

1.09

Activations Density 0.043%