INDEX

Explanations

focusing on

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

さと

1.39

和

1.37

および

1.31

性和

1.23

െങ്കിലും

1.23

さ

1.23

性と

1.21

和大

1.20

roidism

1.18

and

1.16

POSITIVE LOGITS

 importantly

1.89

 subsequently

1.37

sequently

1.35

 أيضًا

1.31

 preferably

1.31

 zároveň

1.30

 كمان

1.27

PGS

1.27

 seterusnya

1.27

 crucially

1.23

Activations Density 0.174%

No Known Activations