INDEX

Explanations

evaluation, configuration, components, repair

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ly

1.78

thc

1.77

iry

1.75

νας

1.74

 Lyon

1.71

يل

1.69

ROW

1.68

 ось

1.66

ns

1.66

%{

1.64

POSITIVE LOGITS

ramient

1.42

 terapé

1.25

ambahan

1.25

 determinación

1.24

 quirúrg

1.23

 trasero

1.20

itamento

1.19

acidad

1.19

ênd

1.19

 dormancy

1.18

Activations Density 0.057%