INDEX

Explanations

departures and related verbs

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

也得

-3.14

⚌

-2.84

 comical

-2.72

-2.67

 cliquer

-2.61

趼

-2.59

厹

-2.55

 overarching

-2.52

 witty

-2.45



-2.44

POSITIVE LOGITS

镦

2.70

 seems

2.66

 Although

2.59

 appears

2.53

 Surprisingly

2.53

 There

2.52

 sooooo

2.50

 believes

2.45

 makes

2.42

 Though

2.38

Activations Density 0.003%

No Known Activations