Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 2366 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.2107. Min Act: -0.0000

Data Index: 3190 (C4 (Web Text))

Max Activating Token Index: 74

Text #1

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.1655. Min Act: -0.0000

Data Index: 72122 (C4 (Web Text))

Max Activating Token Index: 640

Text #2

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.1090. Min Act: -0.0000

Data Index: 438553 (C4 (Web Text))

Max Activating Token Index: 916

Text #3

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.0830. Min Act: -0.0000

Data Index: 519013 (C4 (Web Text))

Max Activating Token Index: 97

Text #4

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.1288. Min Act: -0.0000

Data Index: 852128 (C4 (Web Text))

Max Activating Token Index: 713

Text #5

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9836. Min Act: -0.0000

Data Index: 498405 (C4 (Web Text))

Max Activating Token Index: 195

Text #6

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.0012. Min Act: -0.0000

Data Index: 776513 (C4 (Web Text))

Max Activating Token Index: 686

Text #7

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9944. Min Act: -0.0000

Data Index: 583096 (C4 (Web Text))

Max Activating Token Index: 356

Text #8

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9229. Min Act: -0.0000

Data Index: 1543212 (Python Code)

Max Activating Token Index: 629

Text #9

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9314. Min Act: -0.0000

Data Index: 1597076 (Python Code)

Max Activating Token Index: 855

Text #10

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9802. Min Act: -0.0000

Data Index: 1103148 (C4 (Web Text))

Max Activating Token Index: 276

Text #11

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9468. Min Act: -0.0000

Data Index: 208055 (C4 (Web Text))

Max Activating Token Index: 157

Text #12

Max Range: 1.2107. Min Range: -1.2107

Max Act: 1.0320. Min Act: -0.0000

Data Index: 1065917 (C4 (Web Text))

Max Activating Token Index: 801

Text #13

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9748. Min Act: -0.0000

Data Index: 89970 (C4 (Web Text))

Max Activating Token Index: 802

Text #14

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9606. Min Act: -0.0000

Data Index: 35635 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9917. Min Act: -0.0000

Data Index: 1579391 (Python Code)

Max Activating Token Index: 369

Text #16

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9181. Min Act: -0.0000

Data Index: 262943 (C4 (Web Text))

Max Activating Token Index: 225

Text #17

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.8550. Min Act: -0.0000

Data Index: 39750 (C4 (Web Text))

Max Activating Token Index: 319

Text #18

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9679. Min Act: -0.0000

Data Index: 9639 (C4 (Web Text))

Max Activating Token Index: 951

Text #19

Max Range: 1.2107. Min Range: -1.2107

Max Act: 0.9711. Min Act: -0.0000

Data Index: 679171 (C4 (Web Text))

Max Activating Token Index: 961