INDEX

Explanations

regression metrics

The neuron detects numeric and statistical tokens—numbers, coefficients, R‐values, parentheses, and other numerical notation in scientific writing.

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

dbus

-0.81

roster

-0.77

 עמ

-0.77

avut

-0.76

hammad

-0.75

eded

-0.74

recognition

-0.73

 dAtA

-0.73

 cân

-0.72

 Rovers

-0.71

POSITIVE LOGITS

2.55

1.98

 coefficient

1.87

 goodness

1.64

 Coefficient

1.63

1.57

 Adjusted

1.39

getR

1.38

Coefficient

1.37

coefficient

1.36

Activations Density 0.017%