INDEX

Explanations

failure or sufficiency

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

ूर

2.58

line

2.47

پرس

2.29

ع

2.29

 باش

2.25

ította

2.25

রূপ

2.21

ゴン

2.20

্ধ্য

2.20

ne

2.20

POSITIVE LOGITS

eret

2.73

ями

2.72

ists

2.65

ውነ

2.64

উন্ড

2.52

<bos>

2.52

\{

2.42

ties

2.40

tements

2.40

isf

2.40

Activations Density 0.217%