Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1794 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.9071. Min Act: -0.0003

Data Index: 1061516 (C4 (Web Text))

Max Activating Token Index: 102

Text #1

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.9402. Min Act: -0.0003

Data Index: 113658 (C4 (Web Text))

Max Activating Token Index: 11

Text #2

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.9249. Min Act: -0.0003

Data Index: 204125 (C4 (Web Text))

Max Activating Token Index: 283

Text #3

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.8370. Min Act: -0.0003

Data Index: 674461 (C4 (Web Text))

Max Activating Token Index: 537

Text #4

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.8332. Min Act: -0.0003

Data Index: 47372 (C4 (Web Text))

Max Activating Token Index: 979

Text #5

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7696. Min Act: -0.0003

Data Index: 211041 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7124. Min Act: -0.0003

Data Index: 814208 (C4 (Web Text))

Max Activating Token Index: 580

Text #7

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7082. Min Act: -0.0003

Data Index: 1114868 (C4 (Web Text))

Max Activating Token Index: 726

Text #8

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7092. Min Act: -0.0003

Data Index: 17235 (C4 (Web Text))

Max Activating Token Index: 24

Text #9

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7468. Min Act: -0.0003

Data Index: 960714 (C4 (Web Text))

Max Activating Token Index: 10

Text #10

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7881. Min Act: -0.0003

Data Index: 55160 (C4 (Web Text))

Max Activating Token Index: 162

Text #11

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7072. Min Act: -0.0003

Data Index: 716871 (C4 (Web Text))

Max Activating Token Index: 620

Text #12

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7623. Min Act: -0.0003

Data Index: 319906 (C4 (Web Text))

Max Activating Token Index: 212

Text #13

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7882. Min Act: -0.0003

Data Index: 737741 (C4 (Web Text))

Max Activating Token Index: 712

Text #14

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.5632. Min Act: -0.0003

Data Index: 1329021 (C4 (Web Text))

Max Activating Token Index: 636

Text #15

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7760. Min Act: -0.0003

Data Index: 588212 (C4 (Web Text))

Max Activating Token Index: 438

Text #16

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7573. Min Act: -0.0003

Data Index: 1355637 (C4 (Web Text))

Max Activating Token Index: 144

Text #17

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7470. Min Act: -0.0003

Data Index: 1303133 (C4 (Web Text))

Max Activating Token Index: 124

Text #18

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7662. Min Act: -0.0003

Data Index: 76525 (C4 (Web Text))

Max Activating Token Index: 901

Text #19

Max Range: 0.9402. Min Range: -0.9402

Max Act: 0.7646. Min Act: -0.0003

Data Index: 794466 (C4 (Web Text))

Max Activating Token Index: 464