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

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

Neuron 768 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.1799. Min Act: -0.0001

Data Index: 1168557 (C4 (Web Text))

Max Activating Token Index: 180

Text #1

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.1761. Min Act: -0.0001

Data Index: 343769 (C4 (Web Text))

Max Activating Token Index: 234

Text #2

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.1853. Min Act: -0.0001

Data Index: 1284041 (C4 (Web Text))

Max Activating Token Index: 185

Text #3

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.3331. Min Act: -0.0001

Data Index: 1294634 (C4 (Web Text))

Max Activating Token Index: 130

Text #4

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.0345. Min Act: -0.0001

Data Index: 281803 (C4 (Web Text))

Max Activating Token Index: 224

Text #5

Max Range: 3.3331. Min Range: -3.3331

Max Act: 3.0453. Min Act: -0.0001

Data Index: 872690 (C4 (Web Text))

Max Activating Token Index: 57

Text #6

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.5927. Min Act: -0.0001

Data Index: 457354 (C4 (Web Text))

Max Activating Token Index: 799

Text #7

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.4549. Min Act: -0.0001

Data Index: 381798 (C4 (Web Text))

Max Activating Token Index: 654

Text #8

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.5566. Min Act: -0.0001

Data Index: 532838 (C4 (Web Text))

Max Activating Token Index: 583

Text #9

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.3700. Min Act: -0.0001

Data Index: 1094875 (C4 (Web Text))

Max Activating Token Index: 636

Text #10

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.4863. Min Act: -0.0001

Data Index: 324827 (C4 (Web Text))

Max Activating Token Index: 482

Text #11

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.3441. Min Act: -0.0001

Data Index: 1322578 (C4 (Web Text))

Max Activating Token Index: 433

Text #12

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.3142. Min Act: -0.0001

Data Index: 1113850 (C4 (Web Text))

Max Activating Token Index: 509

Text #13

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.4289. Min Act: -0.0001

Data Index: 783019 (C4 (Web Text))

Max Activating Token Index: 981

Text #14

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.4998. Min Act: -0.0001

Data Index: 1153273 (C4 (Web Text))

Max Activating Token Index: 408

Text #15

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.6266. Min Act: -0.0001

Data Index: 457837 (C4 (Web Text))

Max Activating Token Index: 549

Text #16

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.2477. Min Act: -0.0001

Data Index: 245966 (C4 (Web Text))

Max Activating Token Index: 404

Text #17

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.3710. Min Act: -0.0001

Data Index: 1293848 (C4 (Web Text))

Max Activating Token Index: 670

Text #18

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.2532. Min Act: -0.0001

Data Index: 316554 (C4 (Web Text))

Max Activating Token Index: 818

Text #19

Max Range: 3.3331. Min Range: -3.3331

Max Act: 2.2182. Min Act: -0.0001

Data Index: 996604 (C4 (Web Text))

Max Activating Token Index: 785