INDEX

Explanations

list items, questions, examples

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

 xổ

1.30

 Слу

1.25

 truyền

1.24

RELAND

1.23

CHROM

1.20

 oreille

1.19

öre

1.16

mless

1.16

ร้าย

1.15

ŏ

1.12

POSITIVE LOGITS

 hali

1.78

 مقارنه

1.65

basically

1.65

过去的

1.64

 مقار

1.61

Comparing

1.59

 exfoli

1.55

占比

1.55

urved

1.55

 sher

1.54

Activations Density 0.003%