INDEX

Explanations

delete from code

The neuron activates specifically on occurrences of the keyword “delete” (including the C++/JavaScript delete operator) in code.

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

༐

-2.02

-1.96

 ihren

-1.76

 ihrer

-1.73

絝

-1.72

 another

-1.67

髎

-1.66

だけど

-1.66

 autre

-1.64

 mentes

-1.60

POSITIVE LOGITS

2.00

 this

1.87

1.69

1.64

There

1.48

what

1.47

{}

1.43

 dikutip

1.42

 kanak

1.42

Activations Density 0.002%