Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3082 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.6708. Min Range: -3.6708

Max Act: 3.6708. Min Act: -0.0000

Data Index: 1006673 (C4 (Web Text))

Max Activating Token Index: 890

Text #1

Max Range: 3.6708. Min Range: -3.6708

Max Act: 3.4331. Min Act: -0.0000

Data Index: 1226321 (C4 (Web Text))

Max Activating Token Index: 972

Text #2

Max Range: 3.6708. Min Range: -3.6708

Max Act: 3.2973. Min Act: -0.0000

Data Index: 20839 (C4 (Web Text))

Max Activating Token Index: 819

Text #3

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.5979. Min Act: -0.0000

Data Index: 475077 (C4 (Web Text))

Max Activating Token Index: 433

Text #4

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.7085. Min Act: -0.0000

Data Index: 772436 (C4 (Web Text))

Max Activating Token Index: 892

Text #5

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.1595. Min Act: -0.0000

Data Index: 1407926 (Python Code)

Max Activating Token Index: 270

Text #6

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.2343. Min Act: -0.0000

Data Index: 737426 (C4 (Web Text))

Max Activating Token Index: 214

Text #7

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.1893. Min Act: -0.0000

Data Index: 1309529 (C4 (Web Text))

Max Activating Token Index: 580

Text #8

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.1899. Min Act: -0.0000

Data Index: 451362 (C4 (Web Text))

Max Activating Token Index: 889

Text #9

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.4057. Min Act: -0.0000

Data Index: 643218 (C4 (Web Text))

Max Activating Token Index: 175

Text #10

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.1306. Min Act: -0.0000

Data Index: 250416 (C4 (Web Text))

Max Activating Token Index: 126

Text #11

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.0100. Min Act: -0.0000

Data Index: 1104055 (C4 (Web Text))

Max Activating Token Index: 880

Text #12

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.1067. Min Act: -0.0000

Data Index: 1542072 (Python Code)

Max Activating Token Index: 312

Text #13

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.9175. Min Act: -0.0000

Data Index: 1025558 (C4 (Web Text))

Max Activating Token Index: 845

Text #14

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.9838. Min Act: -0.0000

Data Index: 1349121 (C4 (Web Text))

Max Activating Token Index: 347

Text #15

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.9551. Min Act: -0.0000

Data Index: 661433 (C4 (Web Text))

Max Activating Token Index: 350

Text #16

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.8297. Min Act: -0.0000

Data Index: 1145964 (C4 (Web Text))

Max Activating Token Index: 990

Text #17

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.8261. Min Act: -0.0000

Data Index: 463108 (C4 (Web Text))

Max Activating Token Index: 816

Text #18

Max Range: 3.6708. Min Range: -3.6708

Max Act: 2.0247. Min Act: -0.0000

Data Index: 454411 (C4 (Web Text))

Max Activating Token Index: 803

Text #19

Max Range: 3.6708. Min Range: -3.6708

Max Act: 1.9137. Min Act: -0.0000

Data Index: 677720 (C4 (Web Text))

Max Activating Token Index: 936