INDEX

Explanations

Reading content

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Read

-0.98

lob

-0.64

 extortion

-0.54

 olduğ

-0.49

Lesen

-0.46

 contemporanea

-0.44

 impossibile

-0.42

räumen

-0.42

 bocca

-0.41

邑

-0.41

POSITIVE LOGITS

LookAnd

1.19

 виправивши

0.85

 للمعارف

0.80

RegistryLite

0.79

 jsPsych

0.78

AccessorTable

0.77

 tartalomajánló

0.77

OwnProperty

0.76

awtextra

0.76

$}}

0.75

Activations Density 0.812%