INDEX

Explanations

trial court

The neuron is specialized to detect the word “trial” (especially in contexts like “trial court”) in legal documents.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits



-2.14

 děti

-1.98

cessed

-1.84

 MessageBoxIcon

-1.83

gencias

-1.81

 Bacillus

-1.80

我们就

-1.77

merkmale

-1.76

sebagai

-1.76

 kteří

-1.76

POSITIVE LOGITS

The

2.53

2.25

1.99

齪

1.98

也是

1.95

1.88

也

1.86

 wiederholt

1.84

In

1.84

By

1.80

Activations Density 0.008%