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

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

Neuron 4512 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.3301. Min Range: -2.3301

Max Act: 2.3301. Min Act: -0.0000

Data Index: 416941 (C4 (Web Text))

Max Activating Token Index: 568

Text #1

Max Range: 2.3301. Min Range: -2.3301

Max Act: 2.0525. Min Act: -0.0000

Data Index: 481366 (C4 (Web Text))

Max Activating Token Index: 849

Text #2

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.9112. Min Act: -0.0000

Data Index: 646393 (C4 (Web Text))

Max Activating Token Index: 333

Text #3

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.8665. Min Act: -0.0000

Data Index: 183464 (C4 (Web Text))

Max Activating Token Index: 443

Text #4

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.8747. Min Act: -0.0000

Data Index: 877974 (C4 (Web Text))

Max Activating Token Index: 242

Text #5

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7577. Min Act: -0.0000

Data Index: 341469 (C4 (Web Text))

Max Activating Token Index: 832

Text #6

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.8026. Min Act: -0.0000

Data Index: 985025 (C4 (Web Text))

Max Activating Token Index: 984

Text #7

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7944. Min Act: -0.0000

Data Index: 1138222 (C4 (Web Text))

Max Activating Token Index: 640

Text #8

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7345. Min Act: -0.0000

Data Index: 1173207 (C4 (Web Text))

Max Activating Token Index: 772

Text #9

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7546. Min Act: -0.0000

Data Index: 989544 (C4 (Web Text))

Max Activating Token Index: 661

Text #10

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7206. Min Act: -0.0000

Data Index: 1464154 (Python Code)

Max Activating Token Index: 49

Text #11

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7470. Min Act: -0.0000

Data Index: 1432191 (Python Code)

Max Activating Token Index: 1009

Text #12

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7007. Min Act: -0.0000

Data Index: 1371689 (Python Code)

Max Activating Token Index: 493

Text #13

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7125. Min Act: -0.0000

Data Index: 1236399 (C4 (Web Text))

Max Activating Token Index: 948

Text #14

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.6260. Min Act: -0.0000

Data Index: 388088 (C4 (Web Text))

Max Activating Token Index: 820

Text #15

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.6701. Min Act: -0.0000

Data Index: 577950 (C4 (Web Text))

Max Activating Token Index: 53

Text #16

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.6637. Min Act: -0.0000

Data Index: 1076003 (C4 (Web Text))

Max Activating Token Index: 559

Text #17

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.7032. Min Act: -0.0000

Data Index: 1381989 (Python Code)

Max Activating Token Index: 463

Text #18

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.8331. Min Act: -0.0000

Data Index: 214967 (C4 (Web Text))

Max Activating Token Index: 894

Text #19

Max Range: 2.3301. Min Range: -2.3301

Max Act: 1.6903. Min Act: -0.0000

Data Index: 121345 (C4 (Web Text))

Max Activating Token Index: 865