Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3100 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.5182. Min Act: -0.0001

Data Index: 1255413 (C4 (Web Text))

Max Activating Token Index: 796

Text #1

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.5103. Min Act: -0.0001

Data Index: 226201 (C4 (Web Text))

Max Activating Token Index: 755

Text #2

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4652. Min Act: -0.0001

Data Index: 93097 (C4 (Web Text))

Max Activating Token Index: 656

Text #3

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4764. Min Act: -0.0001

Data Index: 231934 (C4 (Web Text))

Max Activating Token Index: 764

Text #4

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4721. Min Act: -0.0001

Data Index: 175527 (C4 (Web Text))

Max Activating Token Index: 710

Text #5

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4819. Min Act: -0.0001

Data Index: 348683 (C4 (Web Text))

Max Activating Token Index: 513

Text #6

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4649. Min Act: -0.0001

Data Index: 537166 (C4 (Web Text))

Max Activating Token Index: 356

Text #7

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4470. Min Act: -0.0001

Data Index: 1092933 (C4 (Web Text))

Max Activating Token Index: 100

Text #8

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4794. Min Act: -0.0001

Data Index: 1019383 (C4 (Web Text))

Max Activating Token Index: 919

Text #9

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4245. Min Act: -0.0001

Data Index: 1247669 (C4 (Web Text))

Max Activating Token Index: 224

Text #10

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4542. Min Act: -0.0001

Data Index: 486364 (C4 (Web Text))

Max Activating Token Index: 219

Text #11

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4534. Min Act: -0.0001

Data Index: 876466 (C4 (Web Text))

Max Activating Token Index: 670

Text #12

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4361. Min Act: -0.0001

Data Index: 91175 (C4 (Web Text))

Max Activating Token Index: 455

Text #13

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4182. Min Act: -0.0001

Data Index: 532231 (C4 (Web Text))

Max Activating Token Index: 461

Text #14

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4319. Min Act: -0.0001

Data Index: 970650 (C4 (Web Text))

Max Activating Token Index: 757

Text #15

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4343. Min Act: -0.0001

Data Index: 2120 (C4 (Web Text))

Max Activating Token Index: 339

Text #16

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4154. Min Act: -0.0001

Data Index: 834738 (C4 (Web Text))

Max Activating Token Index: 454

Text #17

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4144. Min Act: -0.0001

Data Index: 1169001 (C4 (Web Text))

Max Activating Token Index: 829

Text #18

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4001. Min Act: -0.0001

Data Index: 987404 (C4 (Web Text))

Max Activating Token Index: 667

Text #19

Max Range: 0.5182. Min Range: -0.5182

Max Act: 0.4251. Min Act: -0.0001

Data Index: 653963 (C4 (Web Text))

Max Activating Token Index: 543