INDEX

Explanations

provided

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aben

-0.27

çĪĨåĩº

-0.26

åħ¶å®ĥéĹ®é¢ĺ

-0.26

 flown

-0.25

esp

-0.24

 learnt

-0.24

 infra

-0.24

åĪĿåĪĽ

-0.24

æīĢéľĢè¦ģçļĦ

-0.24

abin

-0.23

POSITIVE LOGITS

ufs

0.27

adays

0.26

annual

0.26

ä¸ĸäºº

0.25

Catch

0.24

 trilogy

0.24

tm

0.24

TM

0.24

TM

0.24

Three

0.24

Activations Density 0.004%