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

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

Neuron 4981 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: 0.9293. Min Range: -0.9293

Max Act: 0.9293. Min Act: -0.0000

Data Index: 856641 (C4 (Web Text))

Max Activating Token Index: 268

Text #1

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.8526. Min Act: -0.0000

Data Index: 1389561 (Python Code)

Max Activating Token Index: 986

Text #2

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.9021. Min Act: -0.0000

Data Index: 388928 (C4 (Web Text))

Max Activating Token Index: 310

Text #3

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.8034. Min Act: -0.0000

Data Index: 1258769 (C4 (Web Text))

Max Activating Token Index: 203

Text #4

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.8021. Min Act: -0.0000

Data Index: 355931 (C4 (Web Text))

Max Activating Token Index: 627

Text #5

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7727. Min Act: -0.0000

Data Index: 1363518 (Python Code)

Max Activating Token Index: 696

Text #6

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.8181. Min Act: -0.0000

Data Index: 114440 (C4 (Web Text))

Max Activating Token Index: 943

Text #7

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7584. Min Act: -0.0000

Data Index: 119415 (C4 (Web Text))

Max Activating Token Index: 341

Text #8

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7484. Min Act: -0.0000

Data Index: 1241687 (C4 (Web Text))

Max Activating Token Index: 465

Text #9

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7591. Min Act: -0.0000

Data Index: 1124928 (C4 (Web Text))

Max Activating Token Index: 762

Text #10

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7260. Min Act: -0.0000

Data Index: 411366 (C4 (Web Text))

Max Activating Token Index: 797

Text #11

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.6979. Min Act: -0.0000

Data Index: 756342 (C4 (Web Text))

Max Activating Token Index: 751

Text #12

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7133. Min Act: -0.0000

Data Index: 637766 (C4 (Web Text))

Max Activating Token Index: 462

Text #13

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7085. Min Act: -0.0000

Data Index: 939326 (C4 (Web Text))

Max Activating Token Index: 453

Text #14

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7179. Min Act: -0.0000

Data Index: 880778 (C4 (Web Text))

Max Activating Token Index: 257

Text #15

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.6964. Min Act: -0.0000

Data Index: 724808 (C4 (Web Text))

Max Activating Token Index: 917

Text #16

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7328. Min Act: -0.0000

Data Index: 1466826 (Python Code)

Max Activating Token Index: 74

Text #17

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7065. Min Act: -0.0000

Data Index: 1069883 (C4 (Web Text))

Max Activating Token Index: 391

Text #18

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.6529. Min Act: -0.0000

Data Index: 525645 (C4 (Web Text))

Max Activating Token Index: 774

Text #19

Max Range: 0.9293. Min Range: -0.9293

Max Act: 0.7327. Min Act: -0.0000

Data Index: 1325182 (C4 (Web Text))

Max Activating Token Index: 375