Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2196 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.5966. Min Act: -0.0002

Data Index: 1460663 (Python Code)

Max Activating Token Index: 521

Text #1

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.5422. Min Act: -0.0002

Data Index: 1377973 (Python Code)

Max Activating Token Index: 434

Text #2

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3802. Min Act: -0.0002

Data Index: 54312 (C4 (Web Text))

Max Activating Token Index: 45

Text #3

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3634. Min Act: -0.0002

Data Index: 1512674 (Python Code)

Max Activating Token Index: 649

Text #4

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3544. Min Act: -0.0002

Data Index: 1138517 (C4 (Web Text))

Max Activating Token Index: 86

Text #5

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3271. Min Act: -0.0002

Data Index: 1150926 (C4 (Web Text))

Max Activating Token Index: 56

Text #6

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3213. Min Act: -0.0002

Data Index: 1619208 (Python Code)

Max Activating Token Index: 586

Text #7

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3269. Min Act: -0.0002

Data Index: 857443 (C4 (Web Text))

Max Activating Token Index: 44

Text #8

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3197. Min Act: -0.0002

Data Index: 692540 (C4 (Web Text))

Max Activating Token Index: 921

Text #9

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3179. Min Act: -0.0002

Data Index: 1499829 (Python Code)

Max Activating Token Index: 423

Text #10

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3165. Min Act: -0.0002

Data Index: 479419 (C4 (Web Text))

Max Activating Token Index: 667

Text #11

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3185. Min Act: -0.0002

Data Index: 884561 (C4 (Web Text))

Max Activating Token Index: 873

Text #12

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3112. Min Act: -0.0002

Data Index: 1361502 (Python Code)

Max Activating Token Index: 185

Text #13

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3126. Min Act: -0.0002

Data Index: 855440 (C4 (Web Text))

Max Activating Token Index: 291

Text #14

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3108. Min Act: -0.0002

Data Index: 317563 (C4 (Web Text))

Max Activating Token Index: 41

Text #15

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3046. Min Act: -0.0002

Data Index: 717699 (C4 (Web Text))

Max Activating Token Index: 247

Text #16

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3030. Min Act: -0.0002

Data Index: 929576 (C4 (Web Text))

Max Activating Token Index: 617

Text #17

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.2987. Min Act: -0.0002

Data Index: 1482907 (Python Code)

Max Activating Token Index: 202

Text #18

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3037. Min Act: -0.0002

Data Index: 125297 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 0.5966. Min Range: -0.5966

Max Act: 0.3074. Min Act: -0.0002

Data Index: 1550627 (Python Code)

Max Activating Token Index: 8