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

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

Neuron 223 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: 3.0683. Min Range: -3.0683

Max Act: 2.8592. Min Act: -0.0001

Data Index: 207787 (C4 (Web Text))

Max Activating Token Index: 1009

Text #1

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.8740. Min Act: -0.0002

Data Index: 823645 (C4 (Web Text))

Max Activating Token Index: 879

Text #2

Max Range: 3.0683. Min Range: -3.0683

Max Act: 3.0683. Min Act: -0.0002

Data Index: 644819 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.7800. Min Act: -0.0001

Data Index: 1521781 (Python Code)

Max Activating Token Index: 897

Text #4

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.6932. Min Act: -0.0002

Data Index: 233376 (C4 (Web Text))

Max Activating Token Index: 692

Text #5

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.5324. Min Act: -0.0001

Data Index: 1108839 (C4 (Web Text))

Max Activating Token Index: 714

Text #6

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.7988. Min Act: -0.0002

Data Index: 1157294 (C4 (Web Text))

Max Activating Token Index: 710

Text #7

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.4314. Min Act: -0.0001

Data Index: 1501174 (Python Code)

Max Activating Token Index: 943

Text #8

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.3993. Min Act: -0.0001

Data Index: 1614616 (Python Code)

Max Activating Token Index: 584

Text #9

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.2581. Min Act: -0.0002

Data Index: 323739 (C4 (Web Text))

Max Activating Token Index: 652

Text #10

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.2237. Min Act: -0.0001

Data Index: 1490839 (Python Code)

Max Activating Token Index: 524

Text #11

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.3322. Min Act: -0.0002

Data Index: 193191 (C4 (Web Text))

Max Activating Token Index: 1020

Text #12

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.4189. Min Act: -0.0001

Data Index: 308266 (C4 (Web Text))

Max Activating Token Index: 498

Text #13

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.0643. Min Act: -0.0001

Data Index: 1504763 (Python Code)

Max Activating Token Index: 984

Text #14

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.3197. Min Act: -0.0002

Data Index: 48850 (C4 (Web Text))

Max Activating Token Index: 900

Text #15

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.3328. Min Act: -0.0001

Data Index: 1633413 (Python Code)

Max Activating Token Index: 463

Text #16

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.1841. Min Act: -0.0002

Data Index: 1132610 (C4 (Web Text))

Max Activating Token Index: 74

Text #17

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.2536. Min Act: -0.0001

Data Index: 1656927 (Python Code)

Max Activating Token Index: 845

Text #18

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.1003. Min Act: -0.0002

Data Index: 539230 (C4 (Web Text))

Max Activating Token Index: 630

Text #19

Max Range: 3.0683. Min Range: -3.0683

Max Act: 2.1601. Min Act: -0.0001

Data Index: 1387381 (Python Code)

Max Activating Token Index: 33