Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4770 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.7552. Min Act: -0.0000

Data Index: 435899 (C4 (Web Text))

Max Activating Token Index: 290

Text #1

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.6350. Min Act: -0.0000

Data Index: 92792 (C4 (Web Text))

Max Activating Token Index: 553

Text #2

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.4853. Min Act: -0.0000

Data Index: 870620 (C4 (Web Text))

Max Activating Token Index: 967

Text #3

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.2375. Min Act: -0.0000

Data Index: 931289 (C4 (Web Text))

Max Activating Token Index: 636

Text #4

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.1696. Min Act: -0.0000

Data Index: 1264624 (C4 (Web Text))

Max Activating Token Index: 946

Text #5

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.0765. Min Act: -0.0000

Data Index: 1222696 (C4 (Web Text))

Max Activating Token Index: 1009

Text #6

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.0646. Min Act: -0.0000

Data Index: 1640082 (Python Code)

Max Activating Token Index: 947

Text #7

Max Range: 2.7552. Min Range: -2.7552

Max Act: 2.0192. Min Act: -0.0000

Data Index: 485451 (C4 (Web Text))

Max Activating Token Index: 377

Text #8

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.9792. Min Act: -0.0000

Data Index: 752902 (C4 (Web Text))

Max Activating Token Index: 522

Text #9

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.9175. Min Act: -0.0000

Data Index: 561601 (C4 (Web Text))

Max Activating Token Index: 720

Text #10

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8093. Min Act: -0.0000

Data Index: 511117 (C4 (Web Text))

Max Activating Token Index: 772

Text #11

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8945. Min Act: -0.0000

Data Index: 881668 (C4 (Web Text))

Max Activating Token Index: 531

Text #12

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8879. Min Act: -0.0000

Data Index: 1028687 (C4 (Web Text))

Max Activating Token Index: 354

Text #13

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8841. Min Act: -0.0000

Data Index: 334460 (C4 (Web Text))

Max Activating Token Index: 688

Text #14

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8669. Min Act: -0.0000

Data Index: 428967 (C4 (Web Text))

Max Activating Token Index: 188

Text #15

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.8077. Min Act: -0.0000

Data Index: 1179445 (C4 (Web Text))

Max Activating Token Index: 956

Text #16

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.7350. Min Act: -0.0000

Data Index: 294884 (C4 (Web Text))

Max Activating Token Index: 858

Text #17

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.7277. Min Act: -0.0000

Data Index: 195508 (C4 (Web Text))

Max Activating Token Index: 974

Text #18

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.6720. Min Act: -0.0000

Data Index: 718114 (C4 (Web Text))

Max Activating Token Index: 399

Text #19

Max Range: 2.7552. Min Range: -2.7552

Max Act: 1.6616. Min Act: -0.0000

Data Index: 746854 (C4 (Web Text))

Max Activating Token Index: 473