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

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

Neuron 461 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.7157. Min Act: -0.1700

Data Index: 1125431 (C4 (Web Text))

Max Activating Token Index: 594

Text #1

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.5592. Min Act: -0.1700

Data Index: 139937 (C4 (Web Text))

Max Activating Token Index: 808

Text #2

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.4338. Min Act: -0.1700

Data Index: 928627 (C4 (Web Text))

Max Activating Token Index: 568

Text #3

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.4338. Min Act: -0.1700

Data Index: 181635 (C4 (Web Text))

Max Activating Token Index: 704

Text #4

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.4024. Min Act: -0.1700

Data Index: 1120115 (C4 (Web Text))

Max Activating Token Index: 588

Text #5

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 501105 (C4 (Web Text))

Max Activating Token Index: 200

Text #6

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 40323 (C4 (Web Text))

Max Activating Token Index: 877

Text #7

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 834848 (C4 (Web Text))

Max Activating Token Index: 700

Text #8

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 797037 (C4 (Web Text))

Max Activating Token Index: 309

Text #9

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 731965 (C4 (Web Text))

Max Activating Token Index: 200

Text #10

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3554. Min Act: -0.1700

Data Index: 543043 (C4 (Web Text))

Max Activating Token Index: 852

Text #11

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.3711. Min Act: -0.1700

Data Index: 1002409 (C4 (Web Text))

Max Activating Token Index: 901

Text #12

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2769. Min Act: -0.1700

Data Index: 970887 (C4 (Web Text))

Max Activating Token Index: 76

Text #13

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2769. Min Act: -0.1700

Data Index: 301033 (C4 (Web Text))

Max Activating Token Index: 122

Text #14

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2769. Min Act: -0.1700

Data Index: 977369 (C4 (Web Text))

Max Activating Token Index: 864

Text #15

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2769. Min Act: -0.1700

Data Index: 454414 (C4 (Web Text))

Max Activating Token Index: 255

Text #16

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2611. Min Act: -0.1700

Data Index: 17288 (C4 (Web Text))

Max Activating Token Index: 603

Text #17

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2611. Min Act: -0.1700

Data Index: 768016 (C4 (Web Text))

Max Activating Token Index: 282

Text #18

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2611. Min Act: -0.1700

Data Index: 411515 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 3.7157. Min Range: -3.7157

Max Act: 3.2297. Min Act: -0.1700

Data Index: 871624 (C4 (Web Text))

Max Activating Token Index: 220