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

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

Neuron 4130 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2863. Min Range: -1.2863

Max Act: 1.2863. Min Act: -0.0001

Data Index: 543131 (C4 (Web Text))

Max Activating Token Index: 794

Text #1

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8766. Min Act: -0.0001

Data Index: 731962 (C4 (Web Text))

Max Activating Token Index: 885

Text #2

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.9568. Min Act: -0.0001

Data Index: 311498 (C4 (Web Text))

Max Activating Token Index: 952

Text #3

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.9857. Min Act: -0.0001

Data Index: 1339515 (C4 (Web Text))

Max Activating Token Index: 569

Text #4

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.9173. Min Act: -0.0001

Data Index: 1348724 (C4 (Web Text))

Max Activating Token Index: 383

Text #5

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.9330. Min Act: -0.0001

Data Index: 1325046 (C4 (Web Text))

Max Activating Token Index: 382

Text #6

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.9066. Min Act: -0.0001

Data Index: 697267 (C4 (Web Text))

Max Activating Token Index: 251

Text #7

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8836. Min Act: -0.0001

Data Index: 600726 (C4 (Web Text))

Max Activating Token Index: 945

Text #8

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7951. Min Act: -0.0001

Data Index: 1327602 (C4 (Web Text))

Max Activating Token Index: 139

Text #9

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8317. Min Act: -0.0001

Data Index: 1216902 (C4 (Web Text))

Max Activating Token Index: 338

Text #10

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8405. Min Act: -0.0001

Data Index: 1623608 (Python Code)

Max Activating Token Index: 791

Text #11

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7998. Min Act: -0.0001

Data Index: 1493016 (Python Code)

Max Activating Token Index: 943

Text #12

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8992. Min Act: -0.0001

Data Index: 1524378 (Python Code)

Max Activating Token Index: 386

Text #13

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8161. Min Act: -0.0001

Data Index: 529425 (C4 (Web Text))

Max Activating Token Index: 869

Text #14

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7904. Min Act: -0.0001

Data Index: 234020 (C4 (Web Text))

Max Activating Token Index: 30

Text #15

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8109. Min Act: -0.0001

Data Index: 324378 (C4 (Web Text))

Max Activating Token Index: 369

Text #16

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.8234. Min Act: -0.0001

Data Index: 244666 (C4 (Web Text))

Max Activating Token Index: 422

Text #17

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7908. Min Act: -0.0001

Data Index: 708055 (C4 (Web Text))

Max Activating Token Index: 629

Text #18

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7659. Min Act: -0.0001

Data Index: 423301 (C4 (Web Text))

Max Activating Token Index: 482

Text #19

Max Range: 1.2863. Min Range: -1.2863

Max Act: 0.7880. Min Act: -0.0001

Data Index: 81123 (C4 (Web Text))

Max Activating Token Index: 21