INDEX

Explanations

from sources or origins

The neuron is triggered by numerical tokens (particularly standalone numbers or number-like strings such as bibliographic “From 4232.0205”) indicating quantitative or reference data.

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

鸚

0.63

ạp

0.62

igraf

0.61

 อยู่

0.61

 سے

0.60

東西

0.59

綘

0.59

 Posteriormente

0.58

的东西

0.57

 اگر

0.57

POSITIVE LOGITS

 afar

2.22

 whence

1.76

 standpoint

1.74

 segi

1.45

 ጋር

1.41

 طریق

1.37

 abroad

1.31

 scratch

1.30

within

1.25

 within

1.22

Activations Density 0.449%