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

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

Neuron 2328 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2908. Min Act: -0.0001

Data Index: 485663 (C4 (Web Text))

Max Activating Token Index: 184

Text #1

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2453. Min Act: -0.0001

Data Index: 992663 (C4 (Web Text))

Max Activating Token Index: 918

Text #2

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2651. Min Act: -0.0001

Data Index: 833003 (C4 (Web Text))

Max Activating Token Index: 221

Text #3

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2547. Min Act: -0.0001

Data Index: 1318279 (C4 (Web Text))

Max Activating Token Index: 994

Text #4

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2790. Min Act: -0.0001

Data Index: 867737 (C4 (Web Text))

Max Activating Token Index: 947

Text #5

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2479. Min Act: -0.0001

Data Index: 100959 (C4 (Web Text))

Max Activating Token Index: 290

Text #6

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2221. Min Act: -0.0001

Data Index: 70893 (C4 (Web Text))

Max Activating Token Index: 756

Text #7

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2301. Min Act: -0.0001

Data Index: 1373180 (Python Code)

Max Activating Token Index: 1004

Text #8

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2282. Min Act: -0.0001

Data Index: 1592993 (Python Code)

Max Activating Token Index: 152

Text #9

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2414. Min Act: -0.0001

Data Index: 1576781 (Python Code)

Max Activating Token Index: 158

Text #10

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2241. Min Act: -0.0001

Data Index: 1562193 (Python Code)

Max Activating Token Index: 183

Text #11

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2432. Min Act: -0.0001

Data Index: 46962 (C4 (Web Text))

Max Activating Token Index: 986

Text #12

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2244. Min Act: -0.0001

Data Index: 510570 (C4 (Web Text))

Max Activating Token Index: 816

Text #13

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2269. Min Act: -0.0001

Data Index: 1235727 (C4 (Web Text))

Max Activating Token Index: 854

Text #14

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2311. Min Act: -0.0001

Data Index: 1510447 (Python Code)

Max Activating Token Index: 933

Text #15

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2068. Min Act: -0.0001

Data Index: 335390 (C4 (Web Text))

Max Activating Token Index: 924

Text #16

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2138. Min Act: -0.0001

Data Index: 543042 (C4 (Web Text))

Max Activating Token Index: 656

Text #17

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2181. Min Act: -0.0001

Data Index: 1504908 (Python Code)

Max Activating Token Index: 313

Text #18

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2231. Min Act: -0.0001

Data Index: 31503 (C4 (Web Text))

Max Activating Token Index: 238

Text #19

Max Range: 0.2908. Min Range: -0.2908

Max Act: 0.2048. Min Act: -0.0001

Data Index: 1291786 (C4 (Web Text))

Max Activating Token Index: 753