INDEX

Explanations

professional communication templates

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

δ

1.43

）』

1.43

Χ

1.38

 μετά

1.38

 στα

1.36

%；

1.35

η

1.34

၊

1.34

âtel

1.33

 වන

1.32

POSITIVE LOGITS

↵↵

4.07

<start_of_image>

2.34

↵↵↵↵

2.03

<h2>

1.96

↵↵↵

1.85

↵↵↵↵↵

1.84

↵↵↵↵↵↵

1.82

↵↵↵↵↵↵↵↵↵↵

1.80

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.76

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.75

Activations Density 0.419%