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

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

Neuron 599 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5342. Min Act: -0.0005

Data Index: 148354 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5255. Min Act: -0.0005

Data Index: 1293554 (C4 (Web Text))

Max Activating Token Index: 896

Text #2

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5132. Min Act: -0.0005

Data Index: 95512 (C4 (Web Text))

Max Activating Token Index: 499

Text #3

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5268. Min Act: -0.0005

Data Index: 388330 (C4 (Web Text))

Max Activating Token Index: 1007

Text #4

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4992. Min Act: -0.0005

Data Index: 1006966 (C4 (Web Text))

Max Activating Token Index: 671

Text #5

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5175. Min Act: -0.0005

Data Index: 495278 (C4 (Web Text))

Max Activating Token Index: 478

Text #6

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5147. Min Act: -0.0005

Data Index: 1344247 (C4 (Web Text))

Max Activating Token Index: 607

Text #7

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4863. Min Act: -0.0005

Data Index: 766137 (C4 (Web Text))

Max Activating Token Index: 856

Text #8

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5003. Min Act: -0.0005

Data Index: 924696 (C4 (Web Text))

Max Activating Token Index: 415

Text #9

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5156. Min Act: -0.0005

Data Index: 292952 (C4 (Web Text))

Max Activating Token Index: 622

Text #10

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5113. Min Act: -0.0005

Data Index: 1283178 (C4 (Web Text))

Max Activating Token Index: 348

Text #11

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4954. Min Act: -0.0005

Data Index: 465904 (C4 (Web Text))

Max Activating Token Index: 28

Text #12

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4600. Min Act: -0.0005

Data Index: 1118163 (C4 (Web Text))

Max Activating Token Index: 145

Text #13

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4854. Min Act: -0.0005

Data Index: 465298 (C4 (Web Text))

Max Activating Token Index: 520

Text #14

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.5013. Min Act: -0.0005

Data Index: 98579 (C4 (Web Text))

Max Activating Token Index: 826

Text #15

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4668. Min Act: -0.0005

Data Index: 98364 (C4 (Web Text))

Max Activating Token Index: 533

Text #16

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4816. Min Act: -0.0005

Data Index: 1147359 (C4 (Web Text))

Max Activating Token Index: 188

Text #17

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4776. Min Act: -0.0005

Data Index: 110674 (C4 (Web Text))

Max Activating Token Index: 201

Text #18

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4746. Min Act: -0.0005

Data Index: 388749 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 0.5342. Min Range: -0.5342

Max Act: 0.4827. Min Act: -0.0005

Data Index: 973574 (C4 (Web Text))

Max Activating Token Index: 646