INDEX

Explanations

keywords and phrases

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵↵↵↵↵↵↵↵↵↵↵↵↵

-0.07

↵↵↵↵↵↵↵

-0.06

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-0.06

ഽ

-0.06

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-0.06

的其他

-0.06

↵↵↵↵↵↵↵↵↵↵↵

-0.06

↵↵↵↵↵↵↵↵

-0.06

ओं

-0.06

 йо

-0.06

POSITIVE LOGITS

 没有

0.07

 disassembled

0.06

 данном

0.06

 where

0.06

 whereby

0.06

 اثر

0.06

 குறைந்த

0.06

 उर्फ

0.06

ﱠ

0.06

original

0.06

Activations Density 0.002%