INDEX

Explanations

closing punctuation like ) or ,

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

すべての

1.86

leak

1.86

ऱ्या

1.80

toadd

1.74

eva

1.71

icing

1.71

Debit

1.70

ファイルの

1.69

ernet

1.67

ऑ

1.67

POSITIVE LOGITS

أ

1.34

)—

1.30

.)

1.24

 lacking

1.23

)–

1.22

ينية

1.18

రో

1.17

 ما

1.17

 چاہیے۔

1.17

,)

1.16

Activations Density 0.123%