INDEX

Explanations

contrasting or complementary pairs

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-1.88

 มา

-1.84

即是

-1.72

took

-1.69

camisetas

-1.68

camisas

-1.67

ごめん

-1.66

故

-1.66

 hinzugefügt

-1.65

mà

-1.63

POSITIVE LOGITS

and

2.38

\%$

2.02

 Arhi

1.91

 with

1.84

izielle

1.75

 JpaRepository

1.73

厲

1.66

НИЙ

1.63

})$.

1.61

tembro

1.60

Activations Density 0.114%