INDEX

Explanations

only part or selected data

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

atus

-0.20

ä¹Ŀå¹´

-0.20

æ²ĵ

-0.19

æĸ°æĹ¶æľŁ

-0.18

 THEM

-0.18

ä¸ĢåĪĨéĴŁ

-0.18

ooled

-0.18

_Instance

-0.18

åĲĦæł·

-0.17

éĤĪ

-0.17

POSITIVE LOGITS

 saja

0.33

éĥ¨åĪĨ

0.33

ãģłãģĳ

0.28

éĥ¨ä»½

0.28

 subset

0.27

èĢĮå·²

0.27

åįķçĭ¬

0.27

 portion

0.25

ä»ħ

0.25

 specific

0.25

Activations Density 0.293%