Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1515 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.3445. Min Act: -0.0002

Data Index: 727537 (C4 (Web Text))

Max Activating Token Index: 677

Text #1

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.3758. Min Act: -0.0002

Data Index: 893709 (C4 (Web Text))

Max Activating Token Index: 839

Text #2

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2842. Min Act: -0.0002

Data Index: 409868 (C4 (Web Text))

Max Activating Token Index: 542

Text #3

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2634. Min Act: -0.0002

Data Index: 1325780 (C4 (Web Text))

Max Activating Token Index: 475

Text #4

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2551. Min Act: -0.0002

Data Index: 1278210 (C4 (Web Text))

Max Activating Token Index: 649

Text #5

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2241. Min Act: -0.0002

Data Index: 576124 (C4 (Web Text))

Max Activating Token Index: 782

Text #6

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2296. Min Act: -0.0002

Data Index: 615365 (C4 (Web Text))

Max Activating Token Index: 33

Text #7

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2479. Min Act: -0.0002

Data Index: 103243 (C4 (Web Text))

Max Activating Token Index: 912

Text #8

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2074. Min Act: -0.0002

Data Index: 702106 (C4 (Web Text))

Max Activating Token Index: 536

Text #9

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2240. Min Act: -0.0002

Data Index: 584023 (C4 (Web Text))

Max Activating Token Index: 784

Text #10

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2084. Min Act: -0.0002

Data Index: 1330758 (C4 (Web Text))

Max Activating Token Index: 360

Text #11

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2157. Min Act: -0.0002

Data Index: 690694 (C4 (Web Text))

Max Activating Token Index: 742

Text #12

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2175. Min Act: -0.0002

Data Index: 844044 (C4 (Web Text))

Max Activating Token Index: 202

Text #13

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.1978. Min Act: -0.0002

Data Index: 447378 (C4 (Web Text))

Max Activating Token Index: 657

Text #14

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2227. Min Act: -0.0002

Data Index: 1113336 (C4 (Web Text))

Max Activating Token Index: 895

Text #15

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.2022. Min Act: -0.0002

Data Index: 1019728 (C4 (Web Text))

Max Activating Token Index: 620

Text #16

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.1883. Min Act: -0.0002

Data Index: 1436292 (Python Code)

Max Activating Token Index: 441

Text #17

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.1871. Min Act: -0.0002

Data Index: 858856 (C4 (Web Text))

Max Activating Token Index: 51

Text #18

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.1866. Min Act: -0.0002

Data Index: 1237525 (C4 (Web Text))

Max Activating Token Index: 623

Text #19

Max Range: 0.3758. Min Range: -0.3758

Max Act: 0.1849. Min Act: -0.0002

Data Index: 798248 (C4 (Web Text))

Max Activating Token Index: 287