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

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

Neuron 2699 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: 0.9404. Min Range: -0.9404

Max Act: 0.8890. Min Act: -0.0001

Data Index: 1595256 (Python Code)

Max Activating Token Index: 115

Text #1

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.9162. Min Act: -0.0001

Data Index: 979526 (C4 (Web Text))

Max Activating Token Index: 310

Text #2

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.9199. Min Act: -0.0001

Data Index: 1526367 (Python Code)

Max Activating Token Index: 736

Text #3

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.9404. Min Act: -0.0001

Data Index: 1389696 (Python Code)

Max Activating Token Index: 111

Text #4

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8857. Min Act: -0.0001

Data Index: 404168 (C4 (Web Text))

Max Activating Token Index: 332

Text #5

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.9013. Min Act: -0.0001

Data Index: 1457320 (Python Code)

Max Activating Token Index: 11

Text #6

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8853. Min Act: -0.0001

Data Index: 1516646 (Python Code)

Max Activating Token Index: 345

Text #7

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8777. Min Act: -0.0001

Data Index: 647934 (C4 (Web Text))

Max Activating Token Index: 311

Text #8

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8595. Min Act: -0.0001

Data Index: 451274 (C4 (Web Text))

Max Activating Token Index: 349

Text #9

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8143. Min Act: -0.0001

Data Index: 1299939 (C4 (Web Text))

Max Activating Token Index: 181

Text #10

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8497. Min Act: -0.0001

Data Index: 1566861 (Python Code)

Max Activating Token Index: 5

Text #11

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8439. Min Act: -0.0001

Data Index: 318358 (C4 (Web Text))

Max Activating Token Index: 930

Text #12

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8454. Min Act: -0.0001

Data Index: 1339574 (C4 (Web Text))

Max Activating Token Index: 974

Text #13

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8210. Min Act: -0.0001

Data Index: 814375 (C4 (Web Text))

Max Activating Token Index: 877

Text #14

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8380. Min Act: -0.0001

Data Index: 1198897 (C4 (Web Text))

Max Activating Token Index: 347

Text #15

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8038. Min Act: -0.0001

Data Index: 1060767 (C4 (Web Text))

Max Activating Token Index: 374

Text #16

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.7809. Min Act: -0.0001

Data Index: 874974 (C4 (Web Text))

Max Activating Token Index: 143

Text #17

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8370. Min Act: -0.0001

Data Index: 389532 (C4 (Web Text))

Max Activating Token Index: 697

Text #18

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8100. Min Act: -0.0001

Data Index: 1485270 (Python Code)

Max Activating Token Index: 644

Text #19

Max Range: 0.9404. Min Range: -0.9404

Max Act: 0.8022. Min Act: -0.0001

Data Index: 1652210 (Python Code)

Max Activating Token Index: 886