INDEX

Explanations

motorcycle and gun models

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 küm

-1.31

 estos

-1.24

ごめん

-1.23

部分が

-1.23

стях

-1.20

 rythme

-1.19

 fehér

-1.19

 calça

-1.18

these

-1.18

PD

-1.18

POSITIVE LOGITS

 then

1.58

now

1.39

>--}}

1.31

to

1.28

so

1.28

RTHOOK

1.27

毗

1.23

циями

1.20

ába

1.20

રા

1.20

Activations Density 0.010%