INDEX

Explanations

mathematical symbols and operators

The neuron detects numeric expressions (digits, decimals, and 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

and

-2.81

&&

-1.45

 покупке

-1.39

/}

-1.36

</h1>

-1.34

甚至

-1.34

 može

-1.29

 nebo

-1.28

而且

-1.27

お子

-1.27

POSITIVE LOGITS

are

1.84

 that

1.75

怱

1.72

のではないでしょうか

1.59

 procra

1.52

のかもしれません

1.52

 operativa

1.41

ktı

1.39

と思われます

1.39

1.38

Activations Density 0.156%