INDEX

Explanations

thor

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

ritis

-0.25

éĵĥ

-0.25

åĵĪå°Ķ

-0.25

orian

-0.24

wid

-0.24

 deton

-0.24

 detr

-0.24

ertil

-0.24

á»į

-0.24

æ»©

-0.23

POSITIVE LOGITS

stown

0.28

åĴĮå¹³

0.28

Gle

0.25

&type

0.25

atedRoute

0.24

stead

0.24

acom

0.24

èĢĲ

0.24

<main

0.24

acic

0.23

Activations Density 0.004%