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

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

Neuron 3645 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.2116. Min Act: -0.0001

Data Index: 313467 (C4 (Web Text))

Max Activating Token Index: 705

Text #1

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.1687. Min Act: -0.0001

Data Index: 1325194 (C4 (Web Text))

Max Activating Token Index: 248

Text #2

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.1414. Min Act: -0.0001

Data Index: 63611 (C4 (Web Text))

Max Activating Token Index: 339

Text #3

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.1260. Min Act: -0.0001

Data Index: 511611 (C4 (Web Text))

Max Activating Token Index: 713

Text #4

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0964. Min Act: -0.0001

Data Index: 1194273 (C4 (Web Text))

Max Activating Token Index: 813

Text #5

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0859. Min Act: -0.0001

Data Index: 699141 (C4 (Web Text))

Max Activating Token Index: 976

Text #6

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0790. Min Act: -0.0001

Data Index: 1196479 (C4 (Web Text))

Max Activating Token Index: 804

Text #7

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0818. Min Act: -0.0001

Data Index: 148955 (C4 (Web Text))

Max Activating Token Index: 627

Text #8

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0842. Min Act: -0.0001

Data Index: 14487 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0865. Min Act: -0.0001

Data Index: 79259 (C4 (Web Text))

Max Activating Token Index: 602

Text #10

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0876. Min Act: -0.0001

Data Index: 1591873 (Python Code)

Max Activating Token Index: 109

Text #11

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0791. Min Act: -0.0001

Data Index: 1417703 (Python Code)

Max Activating Token Index: 384

Text #12

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0844. Min Act: -0.0001

Data Index: 214137 (C4 (Web Text))

Max Activating Token Index: 593

Text #13

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0868. Min Act: -0.0001

Data Index: 438453 (C4 (Web Text))

Max Activating Token Index: 667

Text #14

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0802. Min Act: -0.0001

Data Index: 1008685 (C4 (Web Text))

Max Activating Token Index: 568

Text #15

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0807. Min Act: -0.0001

Data Index: 849760 (C4 (Web Text))

Max Activating Token Index: 124

Text #16

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0730. Min Act: -0.0001

Data Index: 1584365 (Python Code)

Max Activating Token Index: 732

Text #17

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0729. Min Act: -0.0001

Data Index: 684193 (C4 (Web Text))

Max Activating Token Index: 244

Text #18

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0717. Min Act: -0.0001

Data Index: 644242 (C4 (Web Text))

Max Activating Token Index: 140

Text #19

Max Range: 0.2116. Min Range: -0.2116

Max Act: 0.0714. Min Act: -0.0001

Data Index: 1347923 (C4 (Web Text))

Max Activating Token Index: 12