INDEX

Explanations

web development, code, or technical documentation

The neuron fires on isolated numeric tokens (i.e. standalone numbers).

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

jazdu

-1.84

瘆

-1.66

＊＊＊＊

-1.66

veu

-1.63

 spise

-1.62

فوائد

-1.62

więcej

-1.60

Fø

-1.59

 déchir

-1.59

Bewertung

-1.59

POSITIVE LOGITS

<h3>

1.93

<h2>

1.91

<blockquote>

1.83

<strong>

1.81

では

1.77

 then

1.77

<h4>

1.71

1.63

だが

1.60

Activations Density 0.002%