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

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

Neuron 1466 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1093. Min Range: -1.1093

Max Act: 1.1093. Min Act: -0.0000

Data Index: 390141 (C4 (Web Text))

Max Activating Token Index: 590

Text #1

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.9970. Min Act: -0.0000

Data Index: 1073921 (C4 (Web Text))

Max Activating Token Index: 568

Text #2

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.8698. Min Act: -0.0000

Data Index: 19463 (C4 (Web Text))

Max Activating Token Index: 859

Text #3

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.8636. Min Act: -0.0000

Data Index: 935449 (C4 (Web Text))

Max Activating Token Index: 938

Text #4

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.8886. Min Act: -0.0000

Data Index: 1226073 (C4 (Web Text))

Max Activating Token Index: 916

Text #5

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.8812. Min Act: -0.0000

Data Index: 1514882 (Python Code)

Max Activating Token Index: 211

Text #6

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7469. Min Act: -0.0000

Data Index: 894423 (C4 (Web Text))

Max Activating Token Index: 458

Text #7

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7972. Min Act: -0.0000

Data Index: 1201888 (C4 (Web Text))

Max Activating Token Index: 551

Text #8

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7508. Min Act: -0.0000

Data Index: 1609163 (Python Code)

Max Activating Token Index: 1017

Text #9

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7886. Min Act: -0.0000

Data Index: 191249 (C4 (Web Text))

Max Activating Token Index: 881

Text #10

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7297. Min Act: -0.0000

Data Index: 483708 (C4 (Web Text))

Max Activating Token Index: 523

Text #11

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.8812. Min Act: -0.0000

Data Index: 1458715 (Python Code)

Max Activating Token Index: 404

Text #12

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7095. Min Act: -0.0000

Data Index: 500945 (C4 (Web Text))

Max Activating Token Index: 459

Text #13

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7537. Min Act: -0.0000

Data Index: 1061639 (C4 (Web Text))

Max Activating Token Index: 974

Text #14

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7446. Min Act: -0.0000

Data Index: 1611846 (Python Code)

Max Activating Token Index: 980

Text #15

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7459. Min Act: -0.0000

Data Index: 231612 (C4 (Web Text))

Max Activating Token Index: 657

Text #16

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7279. Min Act: -0.0000

Data Index: 1230954 (C4 (Web Text))

Max Activating Token Index: 302

Text #17

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7372. Min Act: -0.0000

Data Index: 1584636 (Python Code)

Max Activating Token Index: 210

Text #18

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.6875. Min Act: -0.0000

Data Index: 408124 (C4 (Web Text))

Max Activating Token Index: 21

Text #19

Max Range: 1.1093. Min Range: -1.1093

Max Act: 0.7140. Min Act: -0.0000

Data Index: 1431143 (Python Code)

Max Activating Token Index: 455