INDEX

Explanations

Python script

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

*)

1.60

 rumoured

1.58

 supposedly

1.54

umā

1.52

){}

1.52

?)

1.51

？）

1.50

.')

1.50

也就是说

1.49

?)

1.49

POSITIVE LOGITS

 Calculus

1.60

 PHYS

1.46

Fundamentals

1.41

 calculus

1.40

 Fundamentals

1.38

 MATH

1.38

fundamentals

1.37

MATH

1.33

 CHEM

1.32

 Lesson

1.25

Activations Density 0.126%