INDEX

Explanations

start of model response

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

for

1.92

to

1.69

in

1.62

 لل

1.61

 throughout

1.57

 within

1.57

 для

1.45

 from

1.45

on

1.44

of

1.43

POSITIVE LOGITS

 shortage

1.55

 alternation

1.49

 번째

1.45

 onslaught

1.41

 aftermath

1.40

 debacle

1.37

市政府

1.37

 interplay

1.34

*)(

1.33

 masterpiece

1.32

Activations Density 0.255%