INDEX

Explanations

more effective outcomes

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 there

-2.02

 that

-1.54

無事に

-1.41

擷

-1.38

There

-1.38

哪怕

-1.36

).

-1.28

しましたが

-1.27

 даже

-1.23

彼らの

-1.23

POSITIVE LOGITS

]+$

1.33

那些

1.31

]");

1.30

 republi

1.29

 também

1.27

 kota

1.26

鞣

1.25

 behandelt

1.24

}?

1.24

gingen

1.23

Activations Density 0.101%