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

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

Neuron 4141 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.5933. Min Range: -4.5933

Max Act: 4.5933. Min Act: -0.0000

Data Index: 1369801 (Python Code)

Max Activating Token Index: 427

Text #1

Max Range: 4.5933. Min Range: -4.5933

Max Act: 3.6344. Min Act: -0.0000

Data Index: 466535 (C4 (Web Text))

Max Activating Token Index: 667

Text #2

Max Range: 4.5933. Min Range: -4.5933

Max Act: 3.6230. Min Act: -0.0000

Data Index: 1404439 (Python Code)

Max Activating Token Index: 410

Text #3

Max Range: 4.5933. Min Range: -4.5933

Max Act: 3.4968. Min Act: -0.0000

Data Index: 1532070 (Python Code)

Max Activating Token Index: 1004

Text #4

Max Range: 4.5933. Min Range: -4.5933

Max Act: 3.2669. Min Act: -0.0000

Data Index: 345425 (C4 (Web Text))

Max Activating Token Index: 1001

Text #5

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.9919. Min Act: -0.0000

Data Index: 1466106 (Python Code)

Max Activating Token Index: 608

Text #6

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.8011. Min Act: -0.0000

Data Index: 1640161 (Python Code)

Max Activating Token Index: 673

Text #7

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.7837. Min Act: -0.0000

Data Index: 767174 (C4 (Web Text))

Max Activating Token Index: 120

Text #8

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.8549. Min Act: -0.0000

Data Index: 1040025 (C4 (Web Text))

Max Activating Token Index: 987

Text #9

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.5790. Min Act: -0.0000

Data Index: 98669 (C4 (Web Text))

Max Activating Token Index: 443

Text #10

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.5487. Min Act: -0.0000

Data Index: 357695 (C4 (Web Text))

Max Activating Token Index: 582

Text #11

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.5686. Min Act: -0.0000

Data Index: 831074 (C4 (Web Text))

Max Activating Token Index: 932

Text #12

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.2769. Min Act: -0.0000

Data Index: 1273814 (C4 (Web Text))

Max Activating Token Index: 235

Text #13

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.1001. Min Act: -0.0000

Data Index: 1635246 (Python Code)

Max Activating Token Index: 900

Text #14

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.1035. Min Act: -0.0000

Data Index: 1444899 (Python Code)

Max Activating Token Index: 234

Text #15

Max Range: 4.5933. Min Range: -4.5933

Max Act: 2.0375. Min Act: -0.0000

Data Index: 243659 (C4 (Web Text))

Max Activating Token Index: 982

Text #16

Max Range: 4.5933. Min Range: -4.5933

Max Act: 1.9739. Min Act: -0.0000

Data Index: 1276852 (C4 (Web Text))

Max Activating Token Index: 912

Text #17

Max Range: 4.5933. Min Range: -4.5933

Max Act: 1.9884. Min Act: -0.0000

Data Index: 1466918 (Python Code)

Max Activating Token Index: 500

Text #18

Max Range: 4.5933. Min Range: -4.5933

Max Act: 1.9424. Min Act: -0.0000

Data Index: 108763 (C4 (Web Text))

Max Activating Token Index: 311

Text #19

Max Range: 4.5933. Min Range: -4.5933

Max Act: 1.8178. Min Act: -0.0000

Data Index: 287929 (C4 (Web Text))

Max Activating Token Index: 238