INDEX

Explanations

concepts and analysis

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

↵

2.58

…

1.58

↵↵↵

1.57

↵↵↵↵↵↵↵↵↵

1.48

↵↵↵↵↵

1.44

.…

1.34

↵↵↵↵↵↵↵↵

1.31

↵↵↵↵↵↵

1.30

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.26

↵↵↵↵↵↵↵

1.25

POSITIVE LOGITS

 sebagainya

1.82

 semelhantes

1.75

 aşağıdaki

1.74

например

1.68

ücksicht

1.62

 mengembangkan

1.61

embangkan

1.58

 menimbulkan

1.58

ispielsweise

1.57

 menjelaskan

1.56

Activations Density 0.504%