INDEX

Explanations

numbers and symbols

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eme

-0.07

IZA

-0.07

åľ°éĿ¢çļĦ

-0.07

æľīä¸ĭåĪĹæĥħå½¢ä¹ĭä¸ĢçļĦ

-0.07

origin

-0.06

éĤ£åı¯æĺ¯

-0.06

eret

-0.06

ä½ľèĢħæīĢæľī

-0.06

ÑĪÑĤÐ°

-0.06

 origin

-0.06

POSITIVE LOGITS

æĽ´æİ¥è¿ĳ

0.09

æĶ¶æķĽ

0.08

 ÎĺÎ±

0.08

 refinement

0.08

 iter

0.08

 closer

0.08

æİ¥è¿ĳ

0.08

 refined

0.07

efined

0.07

 Iter

0.07

Activations Density 0.002%