INDEX

Explanations

command-line instructions

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

僉

2.16

 sucedido

2.01

 xạ

1.87

 Συ

1.82

שא

1.81

잖아

1.80

ියාව

1.79

 tiến

1.77

 עצ

1.77

𝙽

1.76

POSITIVE LOGITS

����

1.90

pura

1.83

complete

1.77

roads

1.73

poison

1.66

dollar

1.65

��

1.63

nbsp

1.63

shed

1.61

mend

1.58

Activations Density 0.000%