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

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

Neuron 3670 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0525. Min Range: -1.0525

Max Act: 1.0525. Min Act: -0.0000

Data Index: 798427 (C4 (Web Text))

Max Activating Token Index: 823

Text #1

Max Range: 1.0525. Min Range: -1.0525

Max Act: 1.0342. Min Act: -0.0000

Data Index: 281946 (C4 (Web Text))

Max Activating Token Index: 23

Text #2

Max Range: 1.0525. Min Range: -1.0525

Max Act: 1.0483. Min Act: -0.0000

Data Index: 342132 (C4 (Web Text))

Max Activating Token Index: 963

Text #3

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.9505. Min Act: -0.0000

Data Index: 156013 (C4 (Web Text))

Max Activating Token Index: 800

Text #4

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8950. Min Act: -0.0000

Data Index: 1035102 (C4 (Web Text))

Max Activating Token Index: 139

Text #5

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8265. Min Act: -0.0000

Data Index: 1278496 (C4 (Web Text))

Max Activating Token Index: 984

Text #6

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8636. Min Act: -0.0000

Data Index: 871763 (C4 (Web Text))

Max Activating Token Index: 569

Text #7

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7684. Min Act: -0.0000

Data Index: 396639 (C4 (Web Text))

Max Activating Token Index: 340

Text #8

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8621. Min Act: -0.0000

Data Index: 55788 (C4 (Web Text))

Max Activating Token Index: 238

Text #9

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7687. Min Act: -0.0000

Data Index: 960210 (C4 (Web Text))

Max Activating Token Index: 782

Text #10

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8212. Min Act: -0.0000

Data Index: 918750 (C4 (Web Text))

Max Activating Token Index: 840

Text #11

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7678. Min Act: -0.0000

Data Index: 679050 (C4 (Web Text))

Max Activating Token Index: 489

Text #12

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7895. Min Act: -0.0000

Data Index: 680777 (C4 (Web Text))

Max Activating Token Index: 852

Text #13

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8885. Min Act: -0.0000

Data Index: 553541 (C4 (Web Text))

Max Activating Token Index: 124

Text #14

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.6969. Min Act: -0.0000

Data Index: 801500 (C4 (Web Text))

Max Activating Token Index: 595

Text #15

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8033. Min Act: -0.0000

Data Index: 361653 (C4 (Web Text))

Max Activating Token Index: 175

Text #16

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8450. Min Act: -0.0000

Data Index: 90092 (C4 (Web Text))

Max Activating Token Index: 302

Text #17

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.8916. Min Act: -0.0000

Data Index: 1153205 (C4 (Web Text))

Max Activating Token Index: 329

Text #18

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7857. Min Act: -0.0000

Data Index: 942368 (C4 (Web Text))

Max Activating Token Index: 805

Text #19

Max Range: 1.0525. Min Range: -1.0525

Max Act: 0.7928. Min Act: -0.0000

Data Index: 576050 (C4 (Web Text))

Max Activating Token Index: 606