INDEX

Explanations

commenting

The neuron fires on instances of the word “comment,” essentially spotting mentions of comments in code or documentation.

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

 leur

-1.68

ﾊ

-1.63

chede

-1.56

 desserts

-1.52

.”

-1.48

 behaupten

-1.44

 somber

-1.43

 desar

-1.41

 své

-1.41

ſu

-1.41

POSITIVE LOGITS

↵

2.17

."));

2.11

.";

2.08

:";

2.00

...";

1.95

</h3>

1.91

 itſelf

1.77

 vasco

1.73

 diablo

1.71

uan

1.70

Activations Density 0.032%