INDEX

Explanations

Gemma team or Cont- prefix

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

firma

1.34

lobulin

1.30

舜

1.28

 settlor

1.26

zyg

1.25

opuerto

1.24

eyer

1.22

梧

1.20

 compuesto

1.19

完

1.16

POSITIVE LOGITS

 Liet

1.12

ってみ

1.10

 hör

1.03

astien

1.01

serv

1.01

guj

1.00

报

1.00

гы

0.96

voje

0.95

roda

0.95

Activations Density 0.119%