INDEX

Explanations

function comments describing code logic

The neuron is activated by standalone symbols and numeric tokens (e.g. “+”, “*”, digits, “%”) typically used for bullets, operators, version numbers, or other formatting in technical text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 সম্পাদ

0.61

%%%%%%%%

0.57

edit

0.53

%%%%%%%%%%%%

0.53

 дослі

0.52

EDIT

0.51

℞

0.50

edited

0.50

改

0.49

istory

0.49

POSITIVE LOGITS

0.79

unf

0.64

\*

0.64

 dead

0.61

∗

0.60

 helt

0.58

 deceased

0.57

మణ

0.57

izers

0.56

$*

0.55

Activations Density 0.000%