Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3491 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1489. Min Act: -0.0001

Data Index: 304498 (C4 (Web Text))

Max Activating Token Index: 423

Text #1

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1385. Min Act: -0.0001

Data Index: 157085 (C4 (Web Text))

Max Activating Token Index: 467

Text #2

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1205. Min Act: -0.0001

Data Index: 216573 (C4 (Web Text))

Max Activating Token Index: 729

Text #3

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1236. Min Act: -0.0002

Data Index: 123236 (C4 (Web Text))

Max Activating Token Index: 506

Text #4

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1222. Min Act: -0.0002

Data Index: 401431 (C4 (Web Text))

Max Activating Token Index: 148

Text #5

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1105. Min Act: -0.0001

Data Index: 883703 (C4 (Web Text))

Max Activating Token Index: 69

Text #6

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1037. Min Act: -0.0001

Data Index: 962185 (C4 (Web Text))

Max Activating Token Index: 467

Text #7

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1068. Min Act: -0.0002

Data Index: 937955 (C4 (Web Text))

Max Activating Token Index: 677

Text #8

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1054. Min Act: -0.0001

Data Index: 407586 (C4 (Web Text))

Max Activating Token Index: 73

Text #9

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1028. Min Act: -0.0002

Data Index: 530609 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1015. Min Act: -0.0001

Data Index: 120089 (C4 (Web Text))

Max Activating Token Index: 837

Text #11

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.1033. Min Act: -0.0001

Data Index: 692148 (C4 (Web Text))

Max Activating Token Index: 927

Text #12

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0941. Min Act: -0.0001

Data Index: 257606 (C4 (Web Text))

Max Activating Token Index: 194

Text #13

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0979. Min Act: -0.0001

Data Index: 553338 (C4 (Web Text))

Max Activating Token Index: 939

Text #14

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0970. Min Act: -0.0001

Data Index: 1075245 (C4 (Web Text))

Max Activating Token Index: 517

Text #15

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0855. Min Act: -0.0001

Data Index: 287547 (C4 (Web Text))

Max Activating Token Index: 47

Text #16

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0915. Min Act: -0.0001

Data Index: 979399 (C4 (Web Text))

Max Activating Token Index: 997

Text #17

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0838. Min Act: -0.0002

Data Index: 269089 (C4 (Web Text))

Max Activating Token Index: 218

Text #18

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0831. Min Act: -0.0002

Data Index: 228961 (C4 (Web Text))

Max Activating Token Index: 769

Text #19

Max Range: 0.1489. Min Range: -0.1489

Max Act: 0.0830. Min Act: -0.0001

Data Index: 4300 (C4 (Web Text))

Max Activating Token Index: 377