INDEX

Explanations

user expressing intent

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ｏ

1.72

 plate

1.63

 thunder

1.61

ｕ

1.51

ｈ

1.47

op

1.47

ｖ

1.44

ვი

1.43

ータル

1.43

absorbing

1.42

POSITIVE LOGITS

Autres

2.04

elijkheid

1.94

ItemID

1.91

此之外

1.84

คคล

1.83

чис

1.81

ChildScrollView

1.81

 scriptlevel

1.80

OMI

1.77

తున్న

1.74

Activations Density 0.009%

No Known Activations