INDEX

Explanations

introductions and explanations

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

}}(

0.96

auté

0.95

(>

0.94

unding

0.91

 प्रतिसाद

0.91

rounded

0.91

蒡

0.90

aut

0.89

rounding

0.88

≤

0.88

POSITIVE LOGITS

 strange

1.60

特別な

1.54

 strangely

1.52

 someth

1.51

 особли

1.49

 khusus

1.48

 Something

1.44

 peculiar

1.42

 особен

1.41

 something

1.37

Activations Density 0.357%