INDEX

Explanations

performing actions on objects

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

但是在

-2.39

因为它

-2.34

-2.08

 οποία

-2.03

但却

-1.88

is

-1.88

<bos>

-1.80

但也

-1.78

-1.77

↵

-1.75

POSITIVE LOGITS

the

4.72

our

3.17

3.05

and

2.27

an

2.19

 both

2.16

plastik

2.14

kredit

2.13

 dámské

2.13

 your

2.09

Activations Density 1.514%