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

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

Neuron 2565 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1287. Min Range: -2.1287

Max Act: 2.1287. Min Act: -0.0000

Data Index: 312181 (C4 (Web Text))

Max Activating Token Index: 414

Text #1

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.8346. Min Act: -0.0000

Data Index: 667647 (C4 (Web Text))

Max Activating Token Index: 256

Text #2

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.9092. Min Act: -0.0000

Data Index: 316304 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.5401. Min Act: -0.0000

Data Index: 558316 (C4 (Web Text))

Max Activating Token Index: 411

Text #4

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.4593. Min Act: -0.0000

Data Index: 463561 (C4 (Web Text))

Max Activating Token Index: 631

Text #5

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.4468. Min Act: -0.0000

Data Index: 438319 (C4 (Web Text))

Max Activating Token Index: 365

Text #6

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.3688. Min Act: -0.0000

Data Index: 1082883 (C4 (Web Text))

Max Activating Token Index: 38

Text #7

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.3351. Min Act: -0.0000

Data Index: 580425 (C4 (Web Text))

Max Activating Token Index: 712

Text #8

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.1471. Min Act: -0.0000

Data Index: 197027 (C4 (Web Text))

Max Activating Token Index: 608

Text #9

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.2090. Min Act: -0.0000

Data Index: 156391 (C4 (Web Text))

Max Activating Token Index: 938

Text #10

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0759. Min Act: -0.0000

Data Index: 817648 (C4 (Web Text))

Max Activating Token Index: 148

Text #11

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0976. Min Act: -0.0000

Data Index: 666619 (C4 (Web Text))

Max Activating Token Index: 60

Text #12

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0875. Min Act: -0.0000

Data Index: 1433846 (Python Code)

Max Activating Token Index: 431

Text #13

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.1428. Min Act: -0.0000

Data Index: 760145 (C4 (Web Text))

Max Activating Token Index: 314

Text #14

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0419. Min Act: -0.0000

Data Index: 158180 (C4 (Web Text))

Max Activating Token Index: 223

Text #15

Max Range: 2.1287. Min Range: -2.1287

Max Act: 0.9748. Min Act: -0.0000

Data Index: 990508 (C4 (Web Text))

Max Activating Token Index: 63

Text #16

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0250. Min Act: -0.0000

Data Index: 469267 (C4 (Web Text))

Max Activating Token Index: 309

Text #17

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0307. Min Act: -0.0000

Data Index: 1653814 (Python Code)

Max Activating Token Index: 449

Text #18

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0882. Min Act: -0.0000

Data Index: 1130326 (C4 (Web Text))

Max Activating Token Index: 87

Text #19

Max Range: 2.1287. Min Range: -2.1287

Max Act: 1.0718. Min Act: -0.0000

Data Index: 1231073 (C4 (Web Text))

Max Activating Token Index: 967