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

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

Neuron 1534 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: 3.0638. Min Range: -3.0638

Max Act: 3.0638. Min Act: -0.0000

Data Index: 1284102 (C4 (Web Text))

Max Activating Token Index: 107

Text #1

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.8428. Min Act: -0.0000

Data Index: 1229181 (C4 (Web Text))

Max Activating Token Index: 676

Text #2

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.8396. Min Act: -0.0000

Data Index: 1058518 (C4 (Web Text))

Max Activating Token Index: 120

Text #3

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.6058. Min Act: -0.0000

Data Index: 703254 (C4 (Web Text))

Max Activating Token Index: 789

Text #4

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.3395. Min Act: -0.0000

Data Index: 1155014 (C4 (Web Text))

Max Activating Token Index: 541

Text #5

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.4452. Min Act: -0.0000

Data Index: 1096838 (C4 (Web Text))

Max Activating Token Index: 700

Text #6

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.3239. Min Act: -0.0000

Data Index: 736601 (C4 (Web Text))

Max Activating Token Index: 482

Text #7

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.2703. Min Act: -0.0000

Data Index: 158098 (C4 (Web Text))

Max Activating Token Index: 473

Text #8

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.2155. Min Act: -0.0000

Data Index: 365476 (C4 (Web Text))

Max Activating Token Index: 866

Text #9

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.3158. Min Act: -0.0000

Data Index: 851022 (C4 (Web Text))

Max Activating Token Index: 74

Text #10

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.3165. Min Act: -0.0000

Data Index: 2184 (C4 (Web Text))

Max Activating Token Index: 896

Text #11

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.2921. Min Act: -0.0000

Data Index: 1222668 (C4 (Web Text))

Max Activating Token Index: 559

Text #12

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.4027. Min Act: -0.0000

Data Index: 1316072 (C4 (Web Text))

Max Activating Token Index: 271

Text #13

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.1599. Min Act: -0.0000

Data Index: 1544784 (Python Code)

Max Activating Token Index: 495

Text #14

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.1915. Min Act: -0.0000

Data Index: 624140 (C4 (Web Text))

Max Activating Token Index: 443

Text #15

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.1781. Min Act: -0.0000

Data Index: 1007200 (C4 (Web Text))

Max Activating Token Index: 39

Text #16

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.0683. Min Act: -0.0000

Data Index: 843915 (C4 (Web Text))

Max Activating Token Index: 136

Text #17

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.0416. Min Act: -0.0000

Data Index: 855096 (C4 (Web Text))

Max Activating Token Index: 657

Text #18

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.1402. Min Act: -0.0000

Data Index: 598092 (C4 (Web Text))

Max Activating Token Index: 360

Text #19

Max Range: 3.0638. Min Range: -3.0638

Max Act: 2.0166. Min Act: -0.0000

Data Index: 803423 (C4 (Web Text))

Max Activating Token Index: 727