INDEX

Explanations

Data output/processing

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Flush

-0.77

 flush

-0.71

новниш

-0.71

flush

-0.66

 découvertes

-0.66

 flushed

-0.65

boutin

-0.65

 مرئيه

-0.64

 виправивши

-0.63

 bezeichneter

-0.63

POSITIVE LOGITS

er

0.70

ation

0.60

AxisAlignment

0.56

ant

0.51

ated

0.50

red

0.48

 mass

0.47

ration

0.47

ence

0.45

ment

0.45

Activations Density 0.030%