INDEX

Explanations

levels, eras, decoding

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

th

2.19

 불구

2.02

オブ

1.91

köz

1.88

ו

1.84

্ম্ম

1.82

렇

1.79

<?=$

1.79

rap

1.77

 preview

1.71

POSITIVE LOGITS

ׁ

2.56

्हा

2.43

বীন্দ্র

2.39

ের

2.39

tım

2.38

फारिश

2.28

skich

2.24

𠃍

2.22

<unused1820>

2.19

 austenite

2.16

Activations Density 0.187%