INDEX

Explanations

jokes with 'because' structure

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

Tyl

1.06

 tokopedia

1.06

:/

1.04

Gj

1.04

 Compress

1.03

 Snapshot

1.03

~!

1.02

]=-

1.01

Qh

1.00

 Kinetic

1.00

POSITIVE LOGITS

,\

0.86

,[

0.86

asserie

0.81

orben

0.81

тового

0.79

‌,

0.78

 тоже

0.76

 жены

0.75

átu

0.75

炒

0.74

Activations Density 0.761%