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

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

Neuron 4967 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1831. Min Act: -0.0001

Data Index: 1225200 (C4 (Web Text))

Max Activating Token Index: 135

Text #1

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1364. Min Act: -0.0001

Data Index: 1289744 (C4 (Web Text))

Max Activating Token Index: 781

Text #2

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1431. Min Act: -0.0001

Data Index: 148608 (C4 (Web Text))

Max Activating Token Index: 23

Text #3

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.0909. Min Act: -0.0001

Data Index: 1203955 (C4 (Web Text))

Max Activating Token Index: 736

Text #4

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1054. Min Act: -0.0001

Data Index: 1419475 (Python Code)

Max Activating Token Index: 817

Text #5

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1166. Min Act: -0.0001

Data Index: 1614201 (Python Code)

Max Activating Token Index: 1018

Text #6

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1075. Min Act: -0.0001

Data Index: 1553089 (Python Code)

Max Activating Token Index: 960

Text #7

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1285. Min Act: -0.0001

Data Index: 284868 (C4 (Web Text))

Max Activating Token Index: 896

Text #8

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1100. Min Act: -0.0001

Data Index: 1378461 (Python Code)

Max Activating Token Index: 931

Text #9

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1267. Min Act: -0.0001

Data Index: 491125 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1117. Min Act: -0.0001

Data Index: 482473 (C4 (Web Text))

Max Activating Token Index: 29

Text #11

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1028. Min Act: -0.0001

Data Index: 347887 (C4 (Web Text))

Max Activating Token Index: 224

Text #12

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1081. Min Act: -0.0001

Data Index: 495858 (C4 (Web Text))

Max Activating Token Index: 83

Text #13

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1110. Min Act: -0.0001

Data Index: 1504260 (Python Code)

Max Activating Token Index: 689

Text #14

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1071. Min Act: -0.0001

Data Index: 1537213 (Python Code)

Max Activating Token Index: 850

Text #15

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1093. Min Act: -0.0001

Data Index: 6669 (C4 (Web Text))

Max Activating Token Index: 475

Text #16

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1055. Min Act: -0.0001

Data Index: 1568188 (Python Code)

Max Activating Token Index: 890

Text #17

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1084. Min Act: -0.0001

Data Index: 1392375 (Python Code)

Max Activating Token Index: 821

Text #18

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1126. Min Act: -0.0001

Data Index: 1554802 (Python Code)

Max Activating Token Index: 901

Text #19

Max Range: 0.1831. Min Range: -0.1831

Max Act: 0.1076. Min Act: -0.0001

Data Index: 963608 (C4 (Web Text))

Max Activating Token Index: 96