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

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

Neuron 1383 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: 0.7325. Min Range: -0.7325

Max Act: 0.7325. Min Act: -0.0000

Data Index: 1369358 (Python Code)

Max Activating Token Index: 549

Text #1

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.7181. Min Act: -0.0000

Data Index: 1373375 (Python Code)

Max Activating Token Index: 161

Text #2

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.6994. Min Act: -0.0000

Data Index: 286339 (C4 (Web Text))

Max Activating Token Index: 834

Text #3

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.6012. Min Act: -0.0000

Data Index: 1389451 (Python Code)

Max Activating Token Index: 789

Text #4

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.5778. Min Act: -0.0000

Data Index: 222389 (C4 (Web Text))

Max Activating Token Index: 650

Text #5

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.5132. Min Act: 0.0000

Data Index: 642809 (C4 (Web Text))

Max Activating Token Index: 807

Text #6

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.5081. Min Act: -0.0000

Data Index: 1595240 (Python Code)

Max Activating Token Index: 766

Text #7

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4850. Min Act: -0.0000

Data Index: 1576638 (Python Code)

Max Activating Token Index: 159

Text #8

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.5048. Min Act: -0.0000

Data Index: 1138941 (C4 (Web Text))

Max Activating Token Index: 773

Text #9

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4867. Min Act: -0.0000

Data Index: 1324643 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.5232. Min Act: -0.0000

Data Index: 1378802 (Python Code)

Max Activating Token Index: 923

Text #11

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4991. Min Act: -0.0000

Data Index: 539888 (C4 (Web Text))

Max Activating Token Index: 644

Text #12

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4591. Min Act: -0.0000

Data Index: 1536672 (Python Code)

Max Activating Token Index: 461

Text #13

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.3939. Min Act: 0.0000

Data Index: 87720 (C4 (Web Text))

Max Activating Token Index: 707

Text #14

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4092. Min Act: -0.0000

Data Index: 551749 (C4 (Web Text))

Max Activating Token Index: 57

Text #15

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4008. Min Act: -0.0000

Data Index: 1298375 (C4 (Web Text))

Max Activating Token Index: 506

Text #16

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4155. Min Act: -0.0000

Data Index: 1493434 (Python Code)

Max Activating Token Index: 614

Text #17

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4300. Min Act: -0.0000

Data Index: 719271 (C4 (Web Text))

Max Activating Token Index: 174

Text #18

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.4115. Min Act: 0.0000

Data Index: 397881 (C4 (Web Text))

Max Activating Token Index: 947

Text #19

Max Range: 0.7325. Min Range: -0.7325

Max Act: 0.3597. Min Act: -0.0000

Data Index: 347164 (C4 (Web Text))

Max Activating Token Index: 649