INDEX

Explanations

museums or code construction

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

Flipper

1.39

FRING

1.36

蓋

1.32

 постав

1.31

 cori

1.31

Valve

1.28

 mitra

1.27

enko

1.25

 Guards

1.24

tattoo

1.23

POSITIVE LOGITS

 شناخت

1.18

etlen

1.07

ernen

1.06

 Más

1.04

}^

1.02

改变

1.01

сход

1.00

 lebih

0.98

 хуже

0.96

eds

0.96

Activations Density 0.031%