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

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

Neuron 1662 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: 0.1371. Min Range: -0.1371

Max Act: 0.1371. Min Act: -0.0001

Data Index: 970370 (C4 (Web Text))

Max Activating Token Index: 52

Text #1

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1211. Min Act: -0.0001

Data Index: 1191497 (C4 (Web Text))

Max Activating Token Index: 58

Text #2

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1092. Min Act: -0.0001

Data Index: 1059924 (C4 (Web Text))

Max Activating Token Index: 924

Text #3

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1105. Min Act: -0.0001

Data Index: 837244 (C4 (Web Text))

Max Activating Token Index: 663

Text #4

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1042. Min Act: -0.0001

Data Index: 1085410 (C4 (Web Text))

Max Activating Token Index: 726

Text #5

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1098. Min Act: -0.0001

Data Index: 68087 (C4 (Web Text))

Max Activating Token Index: 967

Text #6

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.1053. Min Act: -0.0001

Data Index: 1291018 (C4 (Web Text))

Max Activating Token Index: 61

Text #7

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0990. Min Act: -0.0001

Data Index: 696157 (C4 (Web Text))

Max Activating Token Index: 26

Text #8

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0985. Min Act: -0.0001

Data Index: 75900 (C4 (Web Text))

Max Activating Token Index: 70

Text #9

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0978. Min Act: -0.0001

Data Index: 990000 (C4 (Web Text))

Max Activating Token Index: 882

Text #10

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0885. Min Act: -0.0001

Data Index: 914650 (C4 (Web Text))

Max Activating Token Index: 201

Text #11

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0925. Min Act: -0.0001

Data Index: 1226514 (C4 (Web Text))

Max Activating Token Index: 647

Text #12

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0905. Min Act: -0.0001

Data Index: 1584823 (Python Code)

Max Activating Token Index: 156

Text #13

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0895. Min Act: -0.0001

Data Index: 1568442 (Python Code)

Max Activating Token Index: 856

Text #14

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0937. Min Act: -0.0001

Data Index: 1344986 (C4 (Web Text))

Max Activating Token Index: 79

Text #15

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0903. Min Act: -0.0001

Data Index: 261584 (C4 (Web Text))

Max Activating Token Index: 407

Text #16

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0897. Min Act: -0.0001

Data Index: 208050 (C4 (Web Text))

Max Activating Token Index: 38

Text #17

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0952. Min Act: -0.0001

Data Index: 280263 (C4 (Web Text))

Max Activating Token Index: 47

Text #18

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0900. Min Act: -0.0001

Data Index: 987230 (C4 (Web Text))

Max Activating Token Index: 1003

Text #19

Max Range: 0.1371. Min Range: -0.1371

Max Act: 0.0944. Min Act: -0.0001

Data Index: 636291 (C4 (Web Text))

Max Activating Token Index: 755