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

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

Neuron 1289 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.8392. Min Act: -0.1692

Data Index: 483708 (C4 (Web Text))

Max Activating Token Index: 325

Text #1

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.6984. Min Act: -0.1697

Data Index: 752052 (C4 (Web Text))

Max Activating Token Index: 184

Text #2

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.6671. Min Act: -0.1699

Data Index: 581691 (C4 (Web Text))

Max Activating Token Index: 195

Text #3

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.6201. Min Act: -0.1700

Data Index: 331324 (C4 (Web Text))

Max Activating Token Index: 989

Text #4

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.5888. Min Act: -0.1700

Data Index: 1309319 (C4 (Web Text))

Max Activating Token Index: 877

Text #5

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4948. Min Act: -0.1700

Data Index: 132383 (C4 (Web Text))

Max Activating Token Index: 785

Text #6

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4635. Min Act: -0.1700

Data Index: 1073343 (C4 (Web Text))

Max Activating Token Index: 1019

Text #7

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4635. Min Act: -0.1700

Data Index: 803391 (C4 (Web Text))

Max Activating Token Index: 132

Text #8

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4635. Min Act: -0.1700

Data Index: 829547 (C4 (Web Text))

Max Activating Token Index: 68

Text #9

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4635. Min Act: -0.1700

Data Index: 453963 (C4 (Web Text))

Max Activating Token Index: 88

Text #10

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4321. Min Act: -0.1688

Data Index: 472813 (C4 (Web Text))

Max Activating Token Index: 925

Text #11

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4478. Min Act: -0.1699

Data Index: 377923 (C4 (Web Text))

Max Activating Token Index: 732

Text #12

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4635. Min Act: -0.1699

Data Index: 1197912 (C4 (Web Text))

Max Activating Token Index: 476

Text #13

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4478. Min Act: -0.1693

Data Index: 176829 (C4 (Web Text))

Max Activating Token Index: 253

Text #14

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4478. Min Act: -0.1700

Data Index: 1279352 (C4 (Web Text))

Max Activating Token Index: 873

Text #15

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4321. Min Act: -0.1700

Data Index: 1068104 (C4 (Web Text))

Max Activating Token Index: 732

Text #16

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4164. Min Act: -0.1697

Data Index: 368619 (C4 (Web Text))

Max Activating Token Index: 191

Text #17

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.4164. Min Act: -0.1700

Data Index: 861780 (C4 (Web Text))

Max Activating Token Index: 670

Text #18

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.3851. Min Act: -0.1700

Data Index: 1453024 (Python Code)

Max Activating Token Index: 119

Text #19

Max Range: 3.8392. Min Range: -3.8392

Max Act: 3.3694. Min Act: -0.1700

Data Index: 579083 (C4 (Web Text))

Max Activating Token Index: 964