INDEX

Explanations

question-answer format

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

alé

-0.76

cete

-0.76

endfor

-0.75

andowski

-0.75

 each

-0.74

ıştır

-0.72

 हूं

-0.71

kedés

-0.71

 μέχρι

-0.70

***********/

-0.69

POSITIVE LOGITS

addAction

1.16

Missouri

0.93

addField

0.92

Puissance

0.85

struts

0.85

getText

0.84

Neuron

0.80

ogn

0.79

 interceptor

0.77

 Conversion

0.76

Activations Density 0.018%