INDEX

Explanations

person speaking or explaining

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

noted

-0.96

styleUrls

-0.93

 않았

-0.90

volve

-0.85

 confirmó

-0.85

 anunció

-0.85

ask

-0.84

 chiedere

-0.82

stated

-0.78

 usize

-0.77

POSITIVE LOGITS

 discuss

3.31

 discusses

3.06

 talk

2.89

 discussing

2.73

 talks

2.67

 explain

2.59

 talked

2.55

 spoke

2.42

 talking

2.38

 discussed

2.38

Activations Density 0.102%