INDEX

Explanations

demonstrative followed by noun instance

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 diplôme

-1.23

 จาก

-1.23

鋅

-1.22

 meub

-1.20

дят

-1.19

 見え

-1.15

símbolo

-1.15

 bidra

-1.14

Garantía

-1.13

鹀

-1.13

POSITIVE LOGITS

 that

1.38

 about

1.35

</h1>

1.30

 this

1.28

</h3>

1.22

)}=

1.21

 these

1.21

在于

1.20

 εδώ

1.18

but

1.16

Activations Density 0.044%

No Known Activations