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

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

Neuron 255 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: 2.9952. Min Range: -2.9952

Max Act: 2.9952. Min Act: -0.0001

Data Index: 392797 (C4 (Web Text))

Max Activating Token Index: 709

Text #1

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6821. Min Act: -0.0001

Data Index: 299855 (C4 (Web Text))

Max Activating Token Index: 77

Text #2

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6396. Min Act: -0.0001

Data Index: 986948 (C4 (Web Text))

Max Activating Token Index: 831

Text #3

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6833. Min Act: -0.0001

Data Index: 158276 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.5202. Min Act: -0.0001

Data Index: 447095 (C4 (Web Text))

Max Activating Token Index: 372

Text #5

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6423. Min Act: -0.0001

Data Index: 1154558 (C4 (Web Text))

Max Activating Token Index: 34

Text #6

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.7688. Min Act: -0.0001

Data Index: 516766 (C4 (Web Text))

Max Activating Token Index: 723

Text #7

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6329. Min Act: -0.0001

Data Index: 1291703 (C4 (Web Text))

Max Activating Token Index: 756

Text #8

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6287. Min Act: -0.0001

Data Index: 722991 (C4 (Web Text))

Max Activating Token Index: 411

Text #9

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.5762. Min Act: -0.0001

Data Index: 518757 (C4 (Web Text))

Max Activating Token Index: 763

Text #10

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.5674. Min Act: -0.0001

Data Index: 370784 (C4 (Web Text))

Max Activating Token Index: 523

Text #11

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.4271. Min Act: -0.0001

Data Index: 109516 (C4 (Web Text))

Max Activating Token Index: 337

Text #12

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.3375. Min Act: -0.0001

Data Index: 1351582 (C4 (Web Text))

Max Activating Token Index: 726

Text #13

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.4225. Min Act: -0.0001

Data Index: 781689 (C4 (Web Text))

Max Activating Token Index: 465

Text #14

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.5675. Min Act: -0.0001

Data Index: 775617 (C4 (Web Text))

Max Activating Token Index: 804

Text #15

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.4282. Min Act: -0.0001

Data Index: 9236 (C4 (Web Text))

Max Activating Token Index: 734

Text #16

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6676. Min Act: -0.0001

Data Index: 421233 (C4 (Web Text))

Max Activating Token Index: 610

Text #17

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.5260. Min Act: -0.0001

Data Index: 915298 (C4 (Web Text))

Max Activating Token Index: 466

Text #18

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6515. Min Act: -0.0001

Data Index: 201731 (C4 (Web Text))

Max Activating Token Index: 493

Text #19

Max Range: 2.9952. Min Range: -2.9952

Max Act: 2.6551. Min Act: -0.0001

Data Index: 403376 (C4 (Web Text))

Max Activating Token Index: 360