INDEX

Explanations

question and numbered lists

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

却不

-1.52

 zowel

-1.50

 time

-1.45

晴れ

-1.41

どちらも

-1.41

な感じ

-1.37

もとも

-1.35

 same

-1.34

これに

-1.34

 그는

-1.33

POSITIVE LOGITS

虽然

2.00

很多

1.98

许多

1.83

 blanda

1.72

 unſ

1.70

︓

1.69

 ſever

1.67

volezza

1.63

我们

1.63

粜

1.61

Activations Density 0.014%