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

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

Neuron 4891 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.6096. Min Act: -0.0000

Data Index: 950701 (C4 (Web Text))

Max Activating Token Index: 974

Text #1

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.6100. Min Act: -0.0000

Data Index: 38949 (C4 (Web Text))

Max Activating Token Index: 267

Text #2

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.5093. Min Act: -0.0000

Data Index: 589318 (C4 (Web Text))

Max Activating Token Index: 355

Text #3

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4953. Min Act: -0.0000

Data Index: 1150135 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4581. Min Act: -0.0000

Data Index: 700508 (C4 (Web Text))

Max Activating Token Index: 959

Text #5

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.5080. Min Act: -0.0000

Data Index: 1565841 (Python Code)

Max Activating Token Index: 614

Text #6

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4087. Min Act: -0.0000

Data Index: 1431094 (Python Code)

Max Activating Token Index: 887

Text #7

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4387. Min Act: -0.0000

Data Index: 726077 (C4 (Web Text))

Max Activating Token Index: 766

Text #8

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4147. Min Act: -0.0000

Data Index: 1438213 (Python Code)

Max Activating Token Index: 601

Text #9

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4154. Min Act: -0.0000

Data Index: 582241 (C4 (Web Text))

Max Activating Token Index: 923

Text #10

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4270. Min Act: -0.0000

Data Index: 1121767 (C4 (Web Text))

Max Activating Token Index: 552

Text #11

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4105. Min Act: -0.0000

Data Index: 1604417 (Python Code)

Max Activating Token Index: 914

Text #12

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4133. Min Act: -0.0000

Data Index: 1016278 (C4 (Web Text))

Max Activating Token Index: 768

Text #13

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3936. Min Act: -0.0000

Data Index: 1209129 (C4 (Web Text))

Max Activating Token Index: 70

Text #14

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.4020. Min Act: -0.0000

Data Index: 936057 (C4 (Web Text))

Max Activating Token Index: 456

Text #15

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3976. Min Act: -0.0000

Data Index: 738469 (C4 (Web Text))

Max Activating Token Index: 386

Text #16

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3640. Min Act: -0.0000

Data Index: 692381 (C4 (Web Text))

Max Activating Token Index: 882

Text #17

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3754. Min Act: -0.0000

Data Index: 1117170 (C4 (Web Text))

Max Activating Token Index: 155

Text #18

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3860. Min Act: -0.0000

Data Index: 1594329 (Python Code)

Max Activating Token Index: 609

Text #19

Max Range: 0.6100. Min Range: -0.6100

Max Act: 0.3834. Min Act: -0.0000

Data Index: 849724 (C4 (Web Text))

Max Activating Token Index: 720