INDEX

Explanations

risk management and exposure

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ΡΑ

-0.81

 religieuse

-0.78

 scap

-0.77

 вспо

-0.76

mények

-0.75

APPING

-0.75

熵

-0.75

 Dordrecht

-0.75

！」

-0.74

 мя

-0.74

POSITIVE LOGITS

 concentration

1.12

 Concentration

1.11

Concentration

1.05

 exposure

1.03

concentration

1.02

 political

1.00

 Exposure

0.99

execution

0.93

exposure

0.92

 Political

0.90

Activations Density 0.011%