INDEX

Explanations

the rank, the probability, the solution

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

 specifics

-0.08

ÑıÐ½

-0.08

olvers

-0.08

 ãĤ¹ãĥĿãĥ³ãĤµãĥ¼ãĤµãĤ¤ãĥĪ

-0.08

ummies

-0.08

agi

-0.07

èĩªå·±ä¹Ł

-0.07

okies

-0.07

asics

-0.07

 Merke

-0.07

POSITIVE LOGITS

aso

0.08

åħ¶ä¸Ńä¹ĭä¸Ģ

0.08

oto

0.07

ed

0.06

ono

0.06

 /**<

0.06

à¸²à¸°

0.06

çĶ¨äºİ

0.06

ence

0.06

ly

0.06

Activations Density 0.047%