Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2017 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0543. Min Act: -0.0004

Data Index: 1097809 (C4 (Web Text))

Max Activating Token Index: 944

Text #1

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0538. Min Act: -0.0003

Data Index: 765593 (C4 (Web Text))

Max Activating Token Index: 227

Text #2

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0508. Min Act: -0.0003

Data Index: 613829 (C4 (Web Text))

Max Activating Token Index: 850

Text #3

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0506. Min Act: -0.0002

Data Index: 641490 (C4 (Web Text))

Max Activating Token Index: 302

Text #4

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0502. Min Act: -0.0003

Data Index: 1003799 (C4 (Web Text))

Max Activating Token Index: 299

Text #5

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0499. Min Act: -0.0004

Data Index: 752554 (C4 (Web Text))

Max Activating Token Index: 263

Text #6

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0495. Min Act: -0.0003

Data Index: 703067 (C4 (Web Text))

Max Activating Token Index: 336

Text #7

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0485. Min Act: -0.0002

Data Index: 44269 (C4 (Web Text))

Max Activating Token Index: 515

Text #8

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0504. Min Act: -0.0003

Data Index: 959121 (C4 (Web Text))

Max Activating Token Index: 184

Text #9

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0500. Min Act: -0.0003

Data Index: 1014314 (C4 (Web Text))

Max Activating Token Index: 421

Text #10

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0496. Min Act: -0.0003

Data Index: 272545 (C4 (Web Text))

Max Activating Token Index: 153

Text #11

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0485. Min Act: -0.0003

Data Index: 672457 (C4 (Web Text))

Max Activating Token Index: 288

Text #12

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0491. Min Act: -0.0003

Data Index: 998520 (C4 (Web Text))

Max Activating Token Index: 157

Text #13

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0484. Min Act: -0.0003

Data Index: 336540 (C4 (Web Text))

Max Activating Token Index: 989

Text #14

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0474. Min Act: -0.0003

Data Index: 846220 (C4 (Web Text))

Max Activating Token Index: 207

Text #15

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0472. Min Act: -0.0003

Data Index: 1246306 (C4 (Web Text))

Max Activating Token Index: 687

Text #16

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0475. Min Act: -0.0003

Data Index: 911222 (C4 (Web Text))

Max Activating Token Index: 870

Text #17

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0484. Min Act: -0.0003

Data Index: 244190 (C4 (Web Text))

Max Activating Token Index: 61

Text #18

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0486. Min Act: -0.0003

Data Index: 943794 (C4 (Web Text))

Max Activating Token Index: 145

Text #19

Max Range: 0.0543. Min Range: -0.0543

Max Act: 0.0478. Min Act: -0.0003

Data Index: 1042133 (C4 (Web Text))

Max Activating Token Index: 310