INDEX

Explanations

The neuron activates on lines made up of repeated punctuation characters used as visual or code‐block separators (e.g. “~~~” or “------”).

Keywords followed by context

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

 consenti

-0.93

 cerim

-0.92

 musul

-0.91

 eccl

-0.91

первых

-0.90

 habari

-0.88

AGR

-0.88

 maisha

-0.88

 pune

-0.87

POSITIVE LOGITS

 also

1.02

 naše

0.98

our

0.90

 your

0.89

 kuten

0.87

 Getränke

0.85

 náš

0.84

 giản

0.84

を目指す

0.83

 també

0.82

Activations Density 0.000%