INDEX

Explanations

separates location from report

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

 également

-1.20

 montaña

-1.13

 което

-1.12

 военной

-1.09

 prachtige

-1.09

 antigüedad

-1.08

 счастье

-1.07

 poliuret

-1.06

 bahía

-1.06

 También

-1.06

POSITIVE LOGITS

It

1.49

The

1.46

 After

1.45

 Despite

1.44

As

1.36

 Although

1.34

 Following

1.34

Following

1.33

With

1.32

 There

1.30

Activations Density 0.051%