INDEX

Explanations

Le/le

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

å·¥ä¸ļåĽŃåĮº

-0.28

NGX

-0.25

moduleId

-0.25

 lying

-0.24

 Cycl

-0.24

oday

-0.23

esen

-0.23

UniqueId

-0.23

 scratches

-0.23

 Finn

-0.23

POSITIVE LOGITS

ait

0.30

iter

0.28

çļĦä¸Ģç§į

0.27

.cont

0.26

avi

0.26

pus

0.26

nameof

0.25

asta

0.25

èļĢ

0.25

catalog

0.24

Activations Density 0.887%