INDEX

Explanations

Canada

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

ä»Ĩ

-0.25

VEN

-0.24

mediate

-0.24

ä»½

-0.23

<*

-0.23

ÑĢÐ¾Ð²

-0.23

å£«

-0.23

igure

-0.23

amba

-0.22

äººåĳĺ

-0.22

POSITIVE LOGITS

@js

0.26

 contra

0.25

Story

0.25

.Transactional

0.25

 Story

0.24

eger

0.24

 orderly

0.24

otron

0.24

ãĤĴèĢĥãģĪ

0.23

witter

0.23

Activations Density 0.049%