INDEX

Explanations

The response appeared

The neuron strongly activates on long, extended spans of text—i.e. unusually lengthy sentences or passages.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Shortly

1.38

 When

1.33

 After

1.33

 Then

1.27

 Tonight

1.27

 During

1.23

 Afterwards

1.23

 Today

1.22

 Yesterday

1.22

After

1.21

POSITIVE LOGITS

 bijvoorbeeld

0.68

 analogues

0.65

 verwenden

0.64

 minify

0.63

 erforderlich

0.63

 monopolist

0.63

 geschikt

0.61

никам

0.61

 yalnızca

0.61

 değildir

0.61

Activations Density 0.110%