INDEX

Explanations

appositive phrases and specific entities

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

erar

3.43

haj

3.15

er

3.13

ી

3.01

erent

2.88

erweise

2.86

hay

2.85

hy

2.85

加坡

2.84

erings

2.83

POSITIVE LOGITS

ृत्व

4.22

ামুটি

3.68

ting

3.56

ৃত্বে

3.36

ravel

3.32

ttt

3.29

rans

3.27

tttt

3.27

ゥ

3.25

्ठा

3.22

Activations Density 3.013%