INDEX

Explanations

word to subword units

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

র

3.36

3.27

ς

3.23

了

3.09

est

3.09

ो

3.01

2.91

ed

2.90

<bos>

2.86

enstein

2.83

POSITIVE LOGITS

hluk

3.09

ঙালি

2.92

ᵑ

2.77

躇

2.61

Појединачна

2.55

 svare

2.54

letos

2.50

๊ก

2.46

ণিজ্য

2.39

anske

2.38

Activations Density 0.001%