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

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

Neuron 3471 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.2938. Min Act: -0.0001

Data Index: 800693 (C4 (Web Text))

Max Activating Token Index: 180

Text #1

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.3037. Min Act: -0.0001

Data Index: 104915 (C4 (Web Text))

Max Activating Token Index: 238

Text #2

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0998. Min Act: -0.0001

Data Index: 694917 (C4 (Web Text))

Max Activating Token Index: 226

Text #3

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.1635. Min Act: -0.0001

Data Index: 1065974 (C4 (Web Text))

Max Activating Token Index: 375

Text #4

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.1516. Min Act: -0.0001

Data Index: 1306873 (C4 (Web Text))

Max Activating Token Index: 273

Text #5

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.2242. Min Act: -0.0001

Data Index: 896082 (C4 (Web Text))

Max Activating Token Index: 850

Text #6

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.1276. Min Act: -0.0001

Data Index: 1321574 (C4 (Web Text))

Max Activating Token Index: 466

Text #7

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.2007. Min Act: -0.0001

Data Index: 530607 (C4 (Web Text))

Max Activating Token Index: 405

Text #8

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.3226. Min Act: -0.0001

Data Index: 250183 (C4 (Web Text))

Max Activating Token Index: 145

Text #9

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.1899. Min Act: -0.0001

Data Index: 538385 (C4 (Web Text))

Max Activating Token Index: 639

Text #10

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0523. Min Act: -0.0001

Data Index: 1052114 (C4 (Web Text))

Max Activating Token Index: 1002

Text #11

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0976. Min Act: -0.0001

Data Index: 437813 (C4 (Web Text))

Max Activating Token Index: 661

Text #12

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.2241. Min Act: -0.0001

Data Index: 250313 (C4 (Web Text))

Max Activating Token Index: 871

Text #13

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0218. Min Act: -0.0001

Data Index: 691044 (C4 (Web Text))

Max Activating Token Index: 312

Text #14

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0445. Min Act: -0.0001

Data Index: 1464816 (Python Code)

Max Activating Token Index: 217

Text #15

Max Range: 1.3226. Min Range: -1.3226

Max Act: 0.9793. Min Act: -0.0001

Data Index: 24825 (C4 (Web Text))

Max Activating Token Index: 563

Text #16

Max Range: 1.3226. Min Range: -1.3226

Max Act: 0.9809. Min Act: -0.0001

Data Index: 937164 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 1.3226. Min Range: -1.3226

Max Act: 1.0269. Min Act: -0.0001

Data Index: 1280517 (C4 (Web Text))

Max Activating Token Index: 782

Text #18

Max Range: 1.3226. Min Range: -1.3226

Max Act: 0.9234. Min Act: -0.0001

Data Index: 324335 (C4 (Web Text))

Max Activating Token Index: 808

Text #19

Max Range: 1.3226. Min Range: -1.3226

Max Act: 0.9447. Min Act: -0.0001

Data Index: 1637779 (Python Code)

Max Activating Token Index: 19