Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 632 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.6496. Min Act: -0.0001

Data Index: 624077 (C4 (Web Text))

Max Activating Token Index: 92

Text #1

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.5200. Min Act: -0.0001

Data Index: 219295 (C4 (Web Text))

Max Activating Token Index: 679

Text #2

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4787. Min Act: -0.0001

Data Index: 1089612 (C4 (Web Text))

Max Activating Token Index: 242

Text #3

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4777. Min Act: -0.0001

Data Index: 445248 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4410. Min Act: -0.0001

Data Index: 795422 (C4 (Web Text))

Max Activating Token Index: 920

Text #5

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4310. Min Act: -0.0001

Data Index: 625483 (C4 (Web Text))

Max Activating Token Index: 888

Text #6

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4623. Min Act: -0.0001

Data Index: 422235 (C4 (Web Text))

Max Activating Token Index: 357

Text #7

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4427. Min Act: -0.0001

Data Index: 985511 (C4 (Web Text))

Max Activating Token Index: 494

Text #8

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4575. Min Act: -0.0001

Data Index: 906078 (C4 (Web Text))

Max Activating Token Index: 57

Text #9

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4420. Min Act: -0.0001

Data Index: 945415 (C4 (Web Text))

Max Activating Token Index: 668

Text #10

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4568. Min Act: -0.0001

Data Index: 150852 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4176. Min Act: -0.0001

Data Index: 794914 (C4 (Web Text))

Max Activating Token Index: 72

Text #12

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4259. Min Act: -0.0001

Data Index: 1161534 (C4 (Web Text))

Max Activating Token Index: 1002

Text #13

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4074. Min Act: -0.0001

Data Index: 696364 (C4 (Web Text))

Max Activating Token Index: 93

Text #14

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4647. Min Act: -0.0001

Data Index: 650988 (C4 (Web Text))

Max Activating Token Index: 341

Text #15

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4177. Min Act: -0.0001

Data Index: 327244 (C4 (Web Text))

Max Activating Token Index: 832

Text #16

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.3996. Min Act: -0.0001

Data Index: 412531 (C4 (Web Text))

Max Activating Token Index: 769

Text #17

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4100. Min Act: -0.0001

Data Index: 1115076 (C4 (Web Text))

Max Activating Token Index: 44

Text #18

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.4024. Min Act: -0.0001

Data Index: 261301 (C4 (Web Text))

Max Activating Token Index: 106

Text #19

Max Range: 0.6496. Min Range: -0.6496

Max Act: 0.3884. Min Act: -0.0001

Data Index: 295046 (C4 (Web Text))

Max Activating Token Index: 115