INDEX

Explanations

identify, find, or mention

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

inge

1.97

oneph

1.93

èche

1.82

 Bever

1.78

čaj

1.76

 Cord

1.75

 Geld

1.75

VMs

1.74

 hedging

1.74

MTV

1.73

POSITIVE LOGITS

 полу

1.90

字体

1.73

поряд

1.63

noe

1.63

 българ

1.57

眉头

1.56

 smelt

1.51

 last

1.49

уга

1.48

这句话

1.47

Activations Density 0.010%