Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1267 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1247. Min Act: -0.0001

Data Index: 1632418 (Python Code)

Max Activating Token Index: 931

Text #1

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1148. Min Act: -0.0001

Data Index: 1349892 (C4 (Web Text))

Max Activating Token Index: 549

Text #2

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1121. Min Act: -0.0001

Data Index: 1574312 (Python Code)

Max Activating Token Index: 844

Text #3

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1052. Min Act: -0.0001

Data Index: 588002 (C4 (Web Text))

Max Activating Token Index: 957

Text #4

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1089. Min Act: -0.0001

Data Index: 1063458 (C4 (Web Text))

Max Activating Token Index: 659

Text #5

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1080. Min Act: -0.0001

Data Index: 401895 (C4 (Web Text))

Max Activating Token Index: 684

Text #6

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1076. Min Act: -0.0001

Data Index: 231194 (C4 (Web Text))

Max Activating Token Index: 695

Text #7

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1112. Min Act: -0.0001

Data Index: 930600 (C4 (Web Text))

Max Activating Token Index: 516

Text #8

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1150. Min Act: -0.0001

Data Index: 110398 (C4 (Web Text))

Max Activating Token Index: 856

Text #9

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1068. Min Act: -0.0001

Data Index: 190970 (C4 (Web Text))

Max Activating Token Index: 631

Text #10

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1024. Min Act: -0.0001

Data Index: 1472041 (Python Code)

Max Activating Token Index: 296

Text #11

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1101. Min Act: -0.0001

Data Index: 568456 (C4 (Web Text))

Max Activating Token Index: 892

Text #12

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1098. Min Act: -0.0001

Data Index: 499408 (C4 (Web Text))

Max Activating Token Index: 816

Text #13

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1082. Min Act: -0.0001

Data Index: 1543851 (Python Code)

Max Activating Token Index: 722

Text #14

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1046. Min Act: -0.0001

Data Index: 124145 (C4 (Web Text))

Max Activating Token Index: 213

Text #15

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1029. Min Act: -0.0001

Data Index: 1621068 (Python Code)

Max Activating Token Index: 769

Text #16

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1064. Min Act: -0.0001

Data Index: 840688 (C4 (Web Text))

Max Activating Token Index: 154

Text #17

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1033. Min Act: -0.0001

Data Index: 864637 (C4 (Web Text))

Max Activating Token Index: 639

Text #18

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1068. Min Act: -0.0001

Data Index: 949656 (C4 (Web Text))

Max Activating Token Index: 869

Text #19

Max Range: 0.1247. Min Range: -0.1247

Max Act: 0.1057. Min Act: -0.0001

Data Index: 48256 (C4 (Web Text))

Max Activating Token Index: 916