INDEX

Explanations

/* comment markers

The neuron responds to words or letter sequences that begin with (and are written in) uppercase letters.

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

-2.19

-2.09

↵↵

-2.08

↵

-2.00

-1.86

 جلوگیری

-1.79

-1.77

 zrobić

-1.70

 uses

-1.61

POSITIVE LOGITS

 impec

1.87

🗹

1.84

of

1.82

 alcuni

1.80

addCriterion

1.73

 estos

1.71

those

1.70

婠

1.70

และ

1.68

ни

1.67

Activations Density 0.000%