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

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

Neuron 3822 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.8955. Min Act: -0.0000

Data Index: 1115204 (C4 (Web Text))

Max Activating Token Index: 718

Text #1

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.7762. Min Act: -0.0000

Data Index: 1178638 (C4 (Web Text))

Max Activating Token Index: 269

Text #2

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.7125. Min Act: -0.0000

Data Index: 60861 (C4 (Web Text))

Max Activating Token Index: 48

Text #3

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6786. Min Act: -0.0000

Data Index: 242217 (C4 (Web Text))

Max Activating Token Index: 223

Text #4

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.7381. Min Act: -0.0000

Data Index: 23553 (C4 (Web Text))

Max Activating Token Index: 621

Text #5

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6872. Min Act: -0.0000

Data Index: 314991 (C4 (Web Text))

Max Activating Token Index: 435

Text #6

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6833. Min Act: -0.0000

Data Index: 83926 (C4 (Web Text))

Max Activating Token Index: 86

Text #7

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6656. Min Act: -0.0000

Data Index: 947530 (C4 (Web Text))

Max Activating Token Index: 412

Text #8

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6273. Min Act: -0.0000

Data Index: 530086 (C4 (Web Text))

Max Activating Token Index: 29

Text #9

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6356. Min Act: -0.0000

Data Index: 180282 (C4 (Web Text))

Max Activating Token Index: 216

Text #10

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6485. Min Act: -0.0000

Data Index: 572838 (C4 (Web Text))

Max Activating Token Index: 997

Text #11

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6082. Min Act: -0.0000

Data Index: 119961 (C4 (Web Text))

Max Activating Token Index: 317

Text #12

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5973. Min Act: -0.0000

Data Index: 1143385 (C4 (Web Text))

Max Activating Token Index: 1005

Text #13

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.6140. Min Act: -0.0000

Data Index: 1559521 (Python Code)

Max Activating Token Index: 741

Text #14

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5682. Min Act: -0.0000

Data Index: 375851 (C4 (Web Text))

Max Activating Token Index: 988

Text #15

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5884. Min Act: -0.0000

Data Index: 102209 (C4 (Web Text))

Max Activating Token Index: 544

Text #16

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5690. Min Act: -0.0000

Data Index: 475106 (C4 (Web Text))

Max Activating Token Index: 732

Text #17

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5999. Min Act: -0.0000

Data Index: 960617 (C4 (Web Text))

Max Activating Token Index: 594

Text #18

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5886. Min Act: -0.0000

Data Index: 774113 (C4 (Web Text))

Max Activating Token Index: 665

Text #19

Max Range: 0.8955. Min Range: -0.8955

Max Act: 0.5441. Min Act: -0.0000

Data Index: 786639 (C4 (Web Text))

Max Activating Token Index: 646