INDEX

Explanations

comparing previous and current changes

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

amen

-0.08

balances

-0.08

bal

-0.07

 balancing

-0.07

tiers

-0.07

eci

-0.07

intang

-0.07

åħ¬åħ±é¢Ħç®Ĺ

-0.07

çĶŁæľº

-0.07

 prospect

-0.07

POSITIVE LOGITS

 continuity

0.10

çĽ¸éĤ»

0.10

 consecutive

0.10

 continuidad

0.10

 prev

0.10

åıĺåĮĸçļĦ

0.10

åºıåĪĹ

0.09

prev

0.09

è¿ŀç»Ń

0.09

 Change

0.09

Activations Density 0.029%