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

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

Neuron 2596 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2463. Min Act: -0.0003

Data Index: 345330 (C4 (Web Text))

Max Activating Token Index: 203

Text #1

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2306. Min Act: -0.0003

Data Index: 840751 (C4 (Web Text))

Max Activating Token Index: 494

Text #2

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2367. Min Act: -0.0003

Data Index: 41793 (C4 (Web Text))

Max Activating Token Index: 43

Text #3

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2400. Min Act: -0.0003

Data Index: 841053 (C4 (Web Text))

Max Activating Token Index: 115

Text #4

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2280. Min Act: -0.0003

Data Index: 363044 (C4 (Web Text))

Max Activating Token Index: 559

Text #5

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2317. Min Act: -0.0003

Data Index: 409336 (C4 (Web Text))

Max Activating Token Index: 64

Text #6

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2101. Min Act: -0.0003

Data Index: 276306 (C4 (Web Text))

Max Activating Token Index: 955

Text #7

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2100. Min Act: -0.0003

Data Index: 949819 (C4 (Web Text))

Max Activating Token Index: 616

Text #8

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2151. Min Act: -0.0003

Data Index: 451137 (C4 (Web Text))

Max Activating Token Index: 441

Text #9

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1913. Min Act: -0.0003

Data Index: 1306087 (C4 (Web Text))

Max Activating Token Index: 758

Text #10

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2036. Min Act: -0.0003

Data Index: 274662 (C4 (Web Text))

Max Activating Token Index: 68

Text #11

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1899. Min Act: -0.0003

Data Index: 930642 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.2024. Min Act: -0.0003

Data Index: 1086506 (C4 (Web Text))

Max Activating Token Index: 188

Text #13

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1868. Min Act: -0.0003

Data Index: 832884 (C4 (Web Text))

Max Activating Token Index: 613

Text #14

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1924. Min Act: -0.0003

Data Index: 781517 (C4 (Web Text))

Max Activating Token Index: 891

Text #15

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1959. Min Act: -0.0003

Data Index: 372524 (C4 (Web Text))

Max Activating Token Index: 232

Text #16

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1930. Min Act: -0.0003

Data Index: 47963 (C4 (Web Text))

Max Activating Token Index: 239

Text #17

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1992. Min Act: -0.0003

Data Index: 1145287 (C4 (Web Text))

Max Activating Token Index: 32

Text #18

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1907. Min Act: -0.0003

Data Index: 1126098 (C4 (Web Text))

Max Activating Token Index: 384

Text #19

Max Range: 0.2463. Min Range: -0.2463

Max Act: 0.1954. Min Act: -0.0003

Data Index: 14007 (C4 (Web Text))

Max Activating Token Index: 56