INDEX

Explanations

you are, the

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

 bleed

1.69

 Tears

1.56

 Forgot

1.54

 Dreaming

1.54

 Naturally

1.53

 enjoyment

1.48

 Assim

1.48

 fortuit

1.46

換

1.44

 reck

1.42

POSITIVE LOGITS

coverageStat

1.66

 पुरालेखित

1.57

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.52

MethodBeat

1.51

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.50

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.50

erkt

1.49

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.47

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.46

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.45

Activations Density 4.636%