INDEX

Explanations

database and code sanitization

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

</h1>

-1.59

</h2>

-1.59

 です

-1.45

If

-1.40

 hätten

-1.35

善良

-1.34

 هناك

-1.33

外界

-1.33

山中

-1.31

 hacía

-1.27

POSITIVE LOGITS

Both

1.76

を買

1.57

 vistazo

1.55

both

1.52

From

1.35

--

1.34

new

1.33

いずれ

1.33

することも

1.30

\_

1.28

Activations Density 0.005%