INDEX

Explanations

introduces definitions and reasons

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

 glorious

1.54

 wonderful

1.52

 magnificent

1.51

 marvellous

1.45

 wondrous

1.38

 vast

1.36

 teeming

1.35

 frontline

1.34

 fantastic

1.33

 immense

1.33

POSITIVE LOGITS

[*][[

1.73

 входит

1.65

 معرف

1.63

 بشكل

1.63

))}

1.61

 тока

1.59

(()

1.58

<<(

1.58

 cạnh

1.57

倸

1.57

Activations Density 0.436%