Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 50 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.0158. Min Range: -1.0158

Max Act: 1.0158. Min Act: -0.0000

Data Index: 420519 (C4 (Web Text))

Max Activating Token Index: 1005

Text #1

Max Range: 1.0158. Min Range: -1.0158

Max Act: 1.0063. Min Act: -0.0001

Data Index: 877564 (C4 (Web Text))

Max Activating Token Index: 775

Text #2

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.8310. Min Act: -0.0000

Data Index: 1122777 (C4 (Web Text))

Max Activating Token Index: 139

Text #3

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.8453. Min Act: -0.0001

Data Index: 555522 (C4 (Web Text))

Max Activating Token Index: 593

Text #4

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.8399. Min Act: -0.0001

Data Index: 1324475 (C4 (Web Text))

Max Activating Token Index: 969

Text #5

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.8190. Min Act: -0.0000

Data Index: 618963 (C4 (Web Text))

Max Activating Token Index: 958

Text #6

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7788. Min Act: -0.0001

Data Index: 1207111 (C4 (Web Text))

Max Activating Token Index: 937

Text #7

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7865. Min Act: -0.0001

Data Index: 141459 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7903. Min Act: -0.0000

Data Index: 196498 (C4 (Web Text))

Max Activating Token Index: 985

Text #9

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.6966. Min Act: -0.0001

Data Index: 244664 (C4 (Web Text))

Max Activating Token Index: 933

Text #10

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7137. Min Act: -0.0001

Data Index: 261138 (C4 (Web Text))

Max Activating Token Index: 979

Text #11

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7420. Min Act: -0.0000

Data Index: 1585645 (Python Code)

Max Activating Token Index: 623

Text #12

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7423. Min Act: -0.0000

Data Index: 1233182 (C4 (Web Text))

Max Activating Token Index: 973

Text #13

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7098. Min Act: -0.0001

Data Index: 1015575 (C4 (Web Text))

Max Activating Token Index: 723

Text #14

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7435. Min Act: -0.0001

Data Index: 1175173 (C4 (Web Text))

Max Activating Token Index: 63

Text #15

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.6854. Min Act: -0.0001

Data Index: 1189267 (C4 (Web Text))

Max Activating Token Index: 562

Text #16

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7024. Min Act: -0.0001

Data Index: 474097 (C4 (Web Text))

Max Activating Token Index: 987

Text #17

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.6943. Min Act: -0.0000

Data Index: 777243 (C4 (Web Text))

Max Activating Token Index: 135

Text #18

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.7169. Min Act: -0.0001

Data Index: 1202631 (C4 (Web Text))

Max Activating Token Index: 544

Text #19

Max Range: 1.0158. Min Range: -1.0158

Max Act: 0.6569. Min Act: -0.0000

Data Index: 386646 (C4 (Web Text))

Max Activating Token Index: 314