Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1105 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5941. Min Act: -0.0001

Data Index: 694386 (C4 (Web Text))

Max Activating Token Index: 83

Text #1

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5690. Min Act: -0.0001

Data Index: 35404 (C4 (Web Text))

Max Activating Token Index: 68

Text #2

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5246. Min Act: -0.0002

Data Index: 259559 (C4 (Web Text))

Max Activating Token Index: 751

Text #3

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5702. Min Act: -0.0001

Data Index: 752809 (C4 (Web Text))

Max Activating Token Index: 366

Text #4

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5625. Min Act: -0.0001

Data Index: 1036206 (C4 (Web Text))

Max Activating Token Index: 78

Text #5

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5998. Min Act: -0.0001

Data Index: 165971 (C4 (Web Text))

Max Activating Token Index: 519

Text #6

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5506. Min Act: -0.0001

Data Index: 393801 (C4 (Web Text))

Max Activating Token Index: 40

Text #7

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5617. Min Act: -0.0001

Data Index: 890397 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5038. Min Act: -0.0002

Data Index: 77967 (C4 (Web Text))

Max Activating Token Index: 596

Text #9

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5732. Min Act: -0.0001

Data Index: 1042042 (C4 (Web Text))

Max Activating Token Index: 215

Text #10

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5333. Min Act: -0.0001

Data Index: 960127 (C4 (Web Text))

Max Activating Token Index: 676

Text #11

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5080. Min Act: -0.0001

Data Index: 531938 (C4 (Web Text))

Max Activating Token Index: 118

Text #12

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4765. Min Act: -0.0001

Data Index: 789627 (C4 (Web Text))

Max Activating Token Index: 69

Text #13

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4439. Min Act: -0.0001

Data Index: 695023 (C4 (Web Text))

Max Activating Token Index: 499

Text #14

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4878. Min Act: -0.0001

Data Index: 535524 (C4 (Web Text))

Max Activating Token Index: 53

Text #15

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5032. Min Act: -0.0001

Data Index: 1293254 (C4 (Web Text))

Max Activating Token Index: 311

Text #16

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.5354. Min Act: -0.0002

Data Index: 432832 (C4 (Web Text))

Max Activating Token Index: 142

Text #17

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4967. Min Act: -0.0002

Data Index: 645080 (C4 (Web Text))

Max Activating Token Index: 126

Text #18

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4854. Min Act: -0.0001

Data Index: 994696 (C4 (Web Text))

Max Activating Token Index: 192

Text #19

Max Range: 0.5998. Min Range: -0.5998

Max Act: 0.4666. Min Act: -0.0001

Data Index: 204667 (C4 (Web Text))

Max Activating Token Index: 678