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

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

Neuron 5111 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.8536. Min Act: -0.0000

Data Index: 697280 (C4 (Web Text))

Max Activating Token Index: 952

Text #1

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.7853. Min Act: -0.0000

Data Index: 1102325 (C4 (Web Text))

Max Activating Token Index: 197

Text #2

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.6949. Min Act: -0.0000

Data Index: 90448 (C4 (Web Text))

Max Activating Token Index: 906

Text #3

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5042. Min Act: -0.0000

Data Index: 972921 (C4 (Web Text))

Max Activating Token Index: 904

Text #4

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.6135. Min Act: -0.0000

Data Index: 1391851 (Python Code)

Max Activating Token Index: 236

Text #5

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.3812. Min Act: -0.0000

Data Index: 142881 (C4 (Web Text))

Max Activating Token Index: 196

Text #6

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5241. Min Act: -0.0000

Data Index: 890001 (C4 (Web Text))

Max Activating Token Index: 407

Text #7

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5870. Min Act: -0.0000

Data Index: 45457 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5008. Min Act: -0.0000

Data Index: 549682 (C4 (Web Text))

Max Activating Token Index: 22

Text #9

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4556. Min Act: -0.0000

Data Index: 1401448 (Python Code)

Max Activating Token Index: 77

Text #10

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4732. Min Act: -0.0000

Data Index: 1513306 (Python Code)

Max Activating Token Index: 875

Text #11

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4701. Min Act: -0.0000

Data Index: 137850 (C4 (Web Text))

Max Activating Token Index: 841

Text #12

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.1773. Min Act: -0.0000

Data Index: 1634463 (Python Code)

Max Activating Token Index: 219

Text #13

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4568. Min Act: -0.0000

Data Index: 1250284 (C4 (Web Text))

Max Activating Token Index: 140

Text #14

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4574. Min Act: -0.0000

Data Index: 980666 (C4 (Web Text))

Max Activating Token Index: 916

Text #15

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.4584. Min Act: -0.0000

Data Index: 30907 (C4 (Web Text))

Max Activating Token Index: 875

Text #16

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5287. Min Act: -0.0000

Data Index: 54405 (C4 (Web Text))

Max Activating Token Index: 162

Text #17

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.2926. Min Act: -0.0000

Data Index: 190226 (C4 (Web Text))

Max Activating Token Index: 303

Text #18

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.5062. Min Act: -0.0000

Data Index: 1605622 (Python Code)

Max Activating Token Index: 93

Text #19

Max Range: 1.8536. Min Range: -1.8536

Max Act: 1.2725. Min Act: -0.0000

Data Index: 1332141 (C4 (Web Text))

Max Activating Token Index: 304