INDEX

Explanations

guess correctly return

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

 manufactures

1.59

 manufacturing

1.56

 manufactured

1.49

 manufacture

1.47

 automotive

1.47

 heating

1.46

 suffers

1.43

 production

1.42

 servicing

1.41

 hauling

1.40

POSITIVE LOGITS

どん

1.21

挑戦

1.17

 கணி

1.12

 глубо

1.11

 пирами

1.11

 சொன்ன

1.11

ปรา

1.10

 schwarzen

1.09

 Literal

1.09

 কাছাকাছি

1.07

Activations Density 0.006%