INDEX

Explanations

lines, paths, and chains

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

 zoude

-1.52

銦

-1.37

 ſta

-1.35

 ジップ

-1.32

 westfalen

-1.30

 Turki

-1.30

actéristi

-1.29



-1.29

 poft

-1.28

 flamengo

-1.27

POSITIVE LOGITS

 This

1.55

 Only

1.47

We

1.45

An

1.43

You

1.40

To

1.36

At

1.35

 Read

1.34

As

1.32

 ALWAYS

1.31

Activations Density 0.136%