Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1274 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0454. Min Act: -0.0002

Data Index: 45106 (C4 (Web Text))

Max Activating Token Index: 743

Text #1

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0471. Min Act: -0.0002

Data Index: 1074864 (C4 (Web Text))

Max Activating Token Index: 538

Text #2

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0440. Min Act: -0.0002

Data Index: 1452171 (Python Code)

Max Activating Token Index: 997

Text #3

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0458. Min Act: -0.0002

Data Index: 394523 (C4 (Web Text))

Max Activating Token Index: 976

Text #4

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0449. Min Act: -0.0002

Data Index: 1600053 (Python Code)

Max Activating Token Index: 281

Text #5

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0443. Min Act: -0.0002

Data Index: 22997 (C4 (Web Text))

Max Activating Token Index: 586

Text #6

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0428. Min Act: -0.0002

Data Index: 499228 (C4 (Web Text))

Max Activating Token Index: 291

Text #7

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0433. Min Act: -0.0002

Data Index: 53098 (C4 (Web Text))

Max Activating Token Index: 647

Text #8

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0422. Min Act: -0.0002

Data Index: 806788 (C4 (Web Text))

Max Activating Token Index: 918

Text #9

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0420. Min Act: -0.0002

Data Index: 840580 (C4 (Web Text))

Max Activating Token Index: 407

Text #10

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0424. Min Act: -0.0002

Data Index: 377311 (C4 (Web Text))

Max Activating Token Index: 199

Text #11

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0428. Min Act: -0.0002

Data Index: 1204240 (C4 (Web Text))

Max Activating Token Index: 950

Text #12

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0423. Min Act: -0.0002

Data Index: 279080 (C4 (Web Text))

Max Activating Token Index: 538

Text #13

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0401. Min Act: -0.0002

Data Index: 715813 (C4 (Web Text))

Max Activating Token Index: 466

Text #14

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0407. Min Act: -0.0002

Data Index: 573312 (C4 (Web Text))

Max Activating Token Index: 210

Text #15

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0399. Min Act: -0.0002

Data Index: 1063886 (C4 (Web Text))

Max Activating Token Index: 813

Text #16

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0410. Min Act: -0.0002

Data Index: 214509 (C4 (Web Text))

Max Activating Token Index: 604

Text #17

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0398. Min Act: -0.0002

Data Index: 1396981 (Python Code)

Max Activating Token Index: 966

Text #18

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0410. Min Act: -0.0002

Data Index: 1001461 (C4 (Web Text))

Max Activating Token Index: 160

Text #19

Max Range: 0.0471. Min Range: -0.0471

Max Act: 0.0390. Min Act: -0.0002

Data Index: 542257 (C4 (Web Text))

Max Activating Token Index: 465