INDEX

Explanations

punctuation followed by a new sentence

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

in

-1.88

</td>

-1.83

-1.78

<td>

-1.66

-1.63

-1.56

こんな感じ

-1.56

↵

-1.55

 måte

-1.55

分か

-1.55

POSITIVE LOGITS

 behulp

1.91

AGUA

1.73

"});

1.73



1.73

・・・」

1.73

鬭

1.70

 acheteurs

1.70

ization

1.69

</strong>

1.68

 quart

1.67

Activations Density 0.005%