INDEX

Explanations

seeking comparisons or specific details

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

<bos>

2.63

${

2.05

setGeometry

1.92

xsd

1.84

diamine

1.78

ς

1.77

 ràng

1.76

茛

1.75

tees

1.74

挛

1.68

POSITIVE LOGITS

మీ

2.33

 overlap

2.03

पूर्ति

1.96

 bursts

1.91

ম

1.88

 dukkh

1.87

 chapel

1.86

APPS

1.86

 overlaps

1.85

礁

1.85

Activations Density 0.445%