INDEX

Explanations

code generation

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

 Vann

0.91

 Bagn

0.87

 examined

0.83

页

0.82

 Glance

0.78

 Bort

0.77

<0xE7>

0.76

 مطالعه

0.75

 examining

0.74

 Horizons

0.72

POSITIVE LOGITS

ALES

1.64

ales

1.58

 reals

1.52

 restocking

1.52

ʲ

1.51

 ProRes

1.48

 conversions

1.46

 preorder

1.44

 prer

1.43

#{

1.43

Activations Density 0.149%