INDEX

Explanations

making processes more efficient

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

2.38

。

1.75

1.67

।

1.59

ambilan

1.43

 sees

1.41

因為

1.41

 because

1.38

因为

1.36

POSITIVE LOGITS

","_

1.42

 lesquelles

1.41

%,

1.38

ActionEvent

1.36

 プラス

1.36

 phấn

1.34

 politiques

1.34

⾏

1.33

ूबर

1.33

`#

1.32

Activations Density 0.119%