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

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

Neuron 4712 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1432. Min Range: -3.1432

Max Act: 3.1432. Min Act: -0.0000

Data Index: 337405 (C4 (Web Text))

Max Activating Token Index: 704

Text #1

Max Range: 3.1432. Min Range: -3.1432

Max Act: 3.1053. Min Act: -0.0000

Data Index: 943669 (C4 (Web Text))

Max Activating Token Index: 401

Text #2

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.8470. Min Act: -0.0000

Data Index: 7077 (C4 (Web Text))

Max Activating Token Index: 733

Text #3

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.5860. Min Act: -0.0000

Data Index: 757703 (C4 (Web Text))

Max Activating Token Index: 533

Text #4

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.8496. Min Act: -0.0000

Data Index: 930136 (C4 (Web Text))

Max Activating Token Index: 980

Text #5

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.6910. Min Act: -0.0000

Data Index: 67850 (C4 (Web Text))

Max Activating Token Index: 100

Text #6

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.4051. Min Act: -0.0000

Data Index: 754296 (C4 (Web Text))

Max Activating Token Index: 617

Text #7

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.2976. Min Act: -0.0000

Data Index: 1317562 (C4 (Web Text))

Max Activating Token Index: 691

Text #8

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.3181. Min Act: -0.0000

Data Index: 1637916 (Python Code)

Max Activating Token Index: 175

Text #9

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8907. Min Act: -0.0000

Data Index: 874392 (C4 (Web Text))

Max Activating Token Index: 879

Text #10

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.9778. Min Act: -0.0000

Data Index: 562813 (C4 (Web Text))

Max Activating Token Index: 379

Text #11

Max Range: 3.1432. Min Range: -3.1432

Max Act: 2.0672. Min Act: -0.0000

Data Index: 1291629 (C4 (Web Text))

Max Activating Token Index: 327

Text #12

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8894. Min Act: -0.0000

Data Index: 1277538 (C4 (Web Text))

Max Activating Token Index: 966

Text #13

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8158. Min Act: -0.0000

Data Index: 1153969 (C4 (Web Text))

Max Activating Token Index: 533

Text #14

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8063. Min Act: -0.0000

Data Index: 357538 (C4 (Web Text))

Max Activating Token Index: 103

Text #15

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8050. Min Act: -0.0000

Data Index: 660816 (C4 (Web Text))

Max Activating Token Index: 286

Text #16

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.8876. Min Act: -0.0000

Data Index: 1374680 (Python Code)

Max Activating Token Index: 645

Text #17

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.6687. Min Act: -0.0000

Data Index: 483866 (C4 (Web Text))

Max Activating Token Index: 816

Text #18

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.6975. Min Act: -0.0000

Data Index: 1301414 (C4 (Web Text))

Max Activating Token Index: 166

Text #19

Max Range: 3.1432. Min Range: -3.1432

Max Act: 1.6332. Min Act: -0.0000

Data Index: 485728 (C4 (Web Text))

Max Activating Token Index: 698