Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1046 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1282. Min Act: -0.0006

Data Index: 273601 (C4 (Web Text))

Max Activating Token Index: 862

Text #1

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1268. Min Act: -0.0006

Data Index: 920001 (C4 (Web Text))

Max Activating Token Index: 64

Text #2

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1137. Min Act: -0.0006

Data Index: 337865 (C4 (Web Text))

Max Activating Token Index: 446

Text #3

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1086. Min Act: -0.0006

Data Index: 1207433 (C4 (Web Text))

Max Activating Token Index: 743

Text #4

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1126. Min Act: -0.0006

Data Index: 372828 (C4 (Web Text))

Max Activating Token Index: 618

Text #5

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1141. Min Act: -0.0005

Data Index: 531539 (C4 (Web Text))

Max Activating Token Index: 236

Text #6

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1139. Min Act: -0.0005

Data Index: 1430954 (Python Code)

Max Activating Token Index: 803

Text #7

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1117. Min Act: -0.0006

Data Index: 1603128 (Python Code)

Max Activating Token Index: 855

Text #8

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1105. Min Act: -0.0006

Data Index: 1413887 (Python Code)

Max Activating Token Index: 190

Text #9

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1060. Min Act: -0.0006

Data Index: 1477327 (Python Code)

Max Activating Token Index: 448

Text #10

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1090. Min Act: -0.0006

Data Index: 814401 (C4 (Web Text))

Max Activating Token Index: 595

Text #11

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1090. Min Act: -0.0006

Data Index: 352176 (C4 (Web Text))

Max Activating Token Index: 918

Text #12

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1084. Min Act: -0.0006

Data Index: 1473567 (Python Code)

Max Activating Token Index: 359

Text #13

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1125. Min Act: -0.0006

Data Index: 417776 (C4 (Web Text))

Max Activating Token Index: 717

Text #14

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1120. Min Act: -0.0006

Data Index: 1236339 (C4 (Web Text))

Max Activating Token Index: 460

Text #15

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.0953. Min Act: -0.0006

Data Index: 1595339 (Python Code)

Max Activating Token Index: 729

Text #16

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1029. Min Act: -0.0006

Data Index: 1356696 (C4 (Web Text))

Max Activating Token Index: 378

Text #17

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1067. Min Act: -0.0006

Data Index: 1454922 (Python Code)

Max Activating Token Index: 600

Text #18

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1025. Min Act: -0.0005

Data Index: 1505379 (Python Code)

Max Activating Token Index: 387

Text #19

Max Range: 0.1282. Min Range: -0.1282

Max Act: 0.1060. Min Act: -0.0006

Data Index: 1629635 (Python Code)

Max Activating Token Index: 804