INDEX

Explanations

error and exception handling

The neuron fires on calls that output or report errors/diagnostics (e.g. MessageBox.Show, Console/println, testing.Fatal, etc.).

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

ով

-0.94

鐺

-0.81

thermia

-0.77

 turbid

-0.74

setResult

-0.74

 Surrender

-0.73

économ

-0.72

arrêt

-0.71

راك

-0.71

 кола

-0.71

POSITIVE LOGITS

Exception

1.48

 Exception

1.20

 exception

1.12

err

1.11

EXCEPTION

1.05

 exceptions

0.96

 error

0.96

exception

0.96

Exceptions

0.93

Error

0.91

Activations Density 0.003%