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

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

Neuron 36 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1693. Min Act: -0.0001

Data Index: 792807 (C4 (Web Text))

Max Activating Token Index: 275

Text #1

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1762. Min Act: -0.0001

Data Index: 882721 (C4 (Web Text))

Max Activating Token Index: 791

Text #2

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1730. Min Act: -0.0001

Data Index: 664737 (C4 (Web Text))

Max Activating Token Index: 753

Text #3

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1593. Min Act: -0.0001

Data Index: 379742 (C4 (Web Text))

Max Activating Token Index: 171

Text #4

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1627. Min Act: -0.0001

Data Index: 86534 (C4 (Web Text))

Max Activating Token Index: 78

Text #5

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1288. Min Act: -0.0001

Data Index: 699221 (C4 (Web Text))

Max Activating Token Index: 952

Text #6

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1327. Min Act: -0.0001

Data Index: 953584 (C4 (Web Text))

Max Activating Token Index: 113

Text #7

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1287. Min Act: -0.0001

Data Index: 1316733 (C4 (Web Text))

Max Activating Token Index: 800

Text #8

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1314. Min Act: -0.0001

Data Index: 401283 (C4 (Web Text))

Max Activating Token Index: 523

Text #9

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1263. Min Act: -0.0001

Data Index: 922431 (C4 (Web Text))

Max Activating Token Index: 57

Text #10

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1262. Min Act: -0.0001

Data Index: 1236603 (C4 (Web Text))

Max Activating Token Index: 220

Text #11

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1202. Min Act: -0.0001

Data Index: 900461 (C4 (Web Text))

Max Activating Token Index: 486

Text #12

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1184. Min Act: -0.0001

Data Index: 1252909 (C4 (Web Text))

Max Activating Token Index: 964

Text #13

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1235. Min Act: -0.0001

Data Index: 248501 (C4 (Web Text))

Max Activating Token Index: 915

Text #14

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1171. Min Act: -0.0001

Data Index: 796669 (C4 (Web Text))

Max Activating Token Index: 396

Text #15

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1109. Min Act: -0.0001

Data Index: 9133 (C4 (Web Text))

Max Activating Token Index: 843

Text #16

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1099. Min Act: -0.0001

Data Index: 901644 (C4 (Web Text))

Max Activating Token Index: 920

Text #17

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1110. Min Act: -0.0001

Data Index: 56450 (C4 (Web Text))

Max Activating Token Index: 45

Text #18

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1067. Min Act: -0.0001

Data Index: 174287 (C4 (Web Text))

Max Activating Token Index: 417

Text #19

Max Range: 0.1762. Min Range: -0.1762

Max Act: 0.1120. Min Act: -0.0001

Data Index: 574402 (C4 (Web Text))

Max Activating Token Index: 685