INDEX

Explanations

contain / contains / containing

The neuron is detecting tokens that contain or consist of numeric characters (i.e. numbers or digits).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

addTo

0.56

removeFrom

0.54

AddTo

0.53

发生在

0.52

 removeFrom

0.51

 ocurr

0.50

 стрем

0.50

 melakukannya

0.50

istič

0.48

 dific

0.48

POSITIVE LOGITS

 containing

3.25

 contains

3.02

 contain

2.94

 Containing

2.83

containing

2.75

contains

2.72

 Contains

2.72

 содержит

2.70

 berisi

2.67

 chứa

2.64

Activations Density 0.736%