INDEX

Explanations

influence how AIfearing it woulddelay purchases expectinglimit trade. They'reCareful Usage: $300

The main thing this neuron does is detect numeric values and statistics (e.g. measurements, percentages, and other numbers).

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

Whilst

2.08

 Whilst

1.86

 whilst

1.63

 içerisinde

1.33

にて

1.32

த்தினை

1.32

 Amongst

1.30

Firstly

1.28

 আমাদেরকে

1.26

様々な

1.21

POSITIVE LOGITS

 staffers

1.22

 amid

1.19

 longtime

1.13

--

1.06

1.03

 मसलन

1.02

—

0.98

``

0.97

 spinoff

0.96

 trove

0.94

Activations Density 0.675%