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

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

Neuron 1948 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.3820. Min Act: -0.0000

Data Index: 427464 (C4 (Web Text))

Max Activating Token Index: 73

Text #1

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.4067. Min Act: -0.0000

Data Index: 919176 (C4 (Web Text))

Max Activating Token Index: 888

Text #2

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.4015. Min Act: -0.0000

Data Index: 1039617 (C4 (Web Text))

Max Activating Token Index: 734

Text #3

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.3857. Min Act: -0.0000

Data Index: 149477 (C4 (Web Text))

Max Activating Token Index: 441

Text #4

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.2587. Min Act: -0.0000

Data Index: 1172384 (C4 (Web Text))

Max Activating Token Index: 586

Text #5

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.2655. Min Act: -0.0000

Data Index: 960183 (C4 (Web Text))

Max Activating Token Index: 692

Text #6

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.2185. Min Act: -0.0000

Data Index: 701898 (C4 (Web Text))

Max Activating Token Index: 770

Text #7

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.2199. Min Act: -0.0000

Data Index: 1455828 (Python Code)

Max Activating Token Index: 270

Text #8

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.2972. Min Act: -0.0000

Data Index: 597359 (C4 (Web Text))

Max Activating Token Index: 195

Text #9

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1511. Min Act: -0.0000

Data Index: 786349 (C4 (Web Text))

Max Activating Token Index: 201

Text #10

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1351. Min Act: -0.0000

Data Index: 967828 (C4 (Web Text))

Max Activating Token Index: 789

Text #11

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1964. Min Act: -0.0000

Data Index: 1579983 (Python Code)

Max Activating Token Index: 363

Text #12

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1126. Min Act: -0.0000

Data Index: 745058 (C4 (Web Text))

Max Activating Token Index: 411

Text #13

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1404. Min Act: -0.0000

Data Index: 538074 (C4 (Web Text))

Max Activating Token Index: 446

Text #14

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1621. Min Act: -0.0000

Data Index: 1192834 (C4 (Web Text))

Max Activating Token Index: 86

Text #15

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1238. Min Act: -0.0000

Data Index: 497799 (C4 (Web Text))

Max Activating Token Index: 392

Text #16

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1191. Min Act: -0.0000

Data Index: 417576 (C4 (Web Text))

Max Activating Token Index: 914

Text #17

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.0829. Min Act: -0.0000

Data Index: 690206 (C4 (Web Text))

Max Activating Token Index: 291

Text #18

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.0488. Min Act: -0.0000

Data Index: 1250197 (C4 (Web Text))

Max Activating Token Index: 828

Text #19

Max Range: 1.4067. Min Range: -1.4067

Max Act: 1.1154. Min Act: -0.0000

Data Index: 577825 (C4 (Web Text))

Max Activating Token Index: 334