INDEX

Explanations

posing a threat or risk

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

 endings

2.82

嗒

2.80

یل

2.75

ോധ

2.74

illée

2.67

一笑

2.67

党

2.58

ات

2.56

ന്യാ

2.52

 expresiones

2.50

POSITIVE LOGITS

ge

2.58

din

2.41

ложи

2.40

}}},

2.24

झा

2.23

sw

2.22

 hoop

2.21

thouse

2.20

fileName

2.16

Cross

2.09

Activations Density 0.012%