INDEX

Explanations

Q and @ markers

The neuron activates on the token “output,” i.e. it detects occurrences of the word “output” in code (filenames or comments).

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

-2.05

We

-2.03

あなたは

-1.80

href

-1.79

-1.75

鹀

-1.73

先日

-1.70

เรา

-1.70

 einem

-1.67

Č

-1.66

POSITIVE LOGITS

 unſ

2.19

齜

2.08

 sombre

2.02

摃

1.98

 ヴィンテージ

1.91

sudadera

1.89



1.88

的吧

1.87

</h3>

1.83

 summarizes

1.81

Activations Density 0.000%