INDEX

Explanations

programming language identifiers

The neuron is mostly triggering on numeric tokens and literals (digits, numbers, or numeric constants) 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

 their

-1.92

on

-1.63

an

-1.62

 from

-1.60

-1.59

 with

-1.56

its

-1.51

 that

-1.48

 both

-1.39

 every

-1.36

POSITIVE LOGITS

で

1.48

↵↵↵↵↵↵↵↵↵↵↵↵↵

1.43

 firs

1.41

 wondrous

1.37

 alcune

1.36

を

1.34

了几

1.32

なので

1.31

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.30

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.29

Activations Density 0.005%