INDEX

Explanations

modern contexts and usage

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

es

3.02

<bos>

2.72

 आलो

2.71

ோ

2.49

இந்நிலையில்

2.43

ubble

2.41

StubCompat

2.39

 belts

2.35

tain

2.34

 scriptlevel

2.33

POSITIVE LOGITS

𝐘

2.84

дентифика

2.64

ما

2.45

 कोई

2.40

ри

2.38

فت

2.38

ر

2.36

ếng

2.33

য়া

2.31

ंझ

2.30

Activations Density 0.001%