INDEX

Explanations

alpha

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

alpha

-1.80

 alpha

-1.54

Alpha

-1.54

beta

-1.52

 Alpha

-1.52

 beta

-1.36

 Beta

-1.20

α

-1.15

Beta

-1.13

ALPHA

-1.09

POSITIVE LOGITS

 myſelf

1.51

Efq

1.48

 itſelf

1.29

 Houſe

1.24

 ―――――

1.23

 faſt

1.23

 houſe

1.22

 ſtate

1.21

ſelf

1.20

 iſt

1.18

Activations Density 1.329%