INDEX

Explanations

annual meetings and conventions

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

are

-1.86

 όταν

-1.67

痦

-1.66

 employ

-1.63

 omdat

-1.62

 exemplify

-1.60

-1.59

 které

-1.57

 preuß

-1.53

POSITIVE LOGITS

鹟

2.05

鲥

2.02

 camiset

1.95



1.91

 parachoque

1.77

of

1.74

ܜ

1.65

𓍊

1.63

🫓

1.59

 vieux

1.59

Activations Density 0.008%