INDEX

Explanations

technical documents

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

haft

-0.30

ilver

-0.27

breadcrumb

-0.26

éĵģè·¯

-0.25

 spur

-0.24

æĲħæĭĮ

-0.23

ENSE

-0.23

íĶ¼

-0.23

 rail

-0.23

_PIPE

-0.23

POSITIVE LOGITS

itore

0.34

ropic

0.27

åı¯è§Ĥ

0.25

oku

0.25

itor

0.25

ä¸Ģä¸ĸ

0.24

 subscriptions

0.24

multipart

0.24

 bunch

0.24

ether

0.24

Activations Density 0.796%