INDEX

Explanations

received, know, username

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

े

1.37

ళి

1.37

ième

1.34

 கருத

1.27

 image

1.26

ellido

1.26

 identification

1.25

essor

1.24

 respectively

1.23

 Deviation

1.23

POSITIVE LOGITS

ofsky

1.53

不会

1.42

좀

1.41

1.40

1.37

몽

1.37

 처음

1.36

 пов

1.35

 ciddi

1.35

 최대한

1.34

Activations Density 0.000%