Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2008 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.5847. Min Act: -0.1700

Data Index: 60933 (C4 (Web Text))

Max Activating Token Index: 325

Text #1

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.4751. Min Act: -0.1700

Data Index: 296186 (C4 (Web Text))

Max Activating Token Index: 923

Text #2

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.4594. Min Act: -0.1700

Data Index: 912838 (C4 (Web Text))

Max Activating Token Index: 612

Text #3

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.4594. Min Act: -0.1700

Data Index: 48555 (C4 (Web Text))

Max Activating Token Index: 618

Text #4

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.4594. Min Act: -0.1700

Data Index: 675224 (C4 (Web Text))

Max Activating Token Index: 655

Text #5

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3653. Min Act: -0.1700

Data Index: 1229478 (C4 (Web Text))

Max Activating Token Index: 792

Text #6

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.4280. Min Act: -0.1700

Data Index: 1232492 (C4 (Web Text))

Max Activating Token Index: 555

Text #7

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3967. Min Act: -0.1700

Data Index: 6894 (C4 (Web Text))

Max Activating Token Index: 493

Text #8

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3967. Min Act: -0.1700

Data Index: 152907 (C4 (Web Text))

Max Activating Token Index: 833

Text #9

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3653. Min Act: -0.1700

Data Index: 575874 (C4 (Web Text))

Max Activating Token Index: 405

Text #10

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3810. Min Act: -0.1700

Data Index: 1325288 (C4 (Web Text))

Max Activating Token Index: 427

Text #11

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3025. Min Act: -0.1700

Data Index: 1250203 (C4 (Web Text))

Max Activating Token Index: 390

Text #12

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.2554. Min Act: -0.1700

Data Index: 788065 (C4 (Web Text))

Max Activating Token Index: 1020

Text #13

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3025. Min Act: -0.1700

Data Index: 862205 (C4 (Web Text))

Max Activating Token Index: 7

Text #14

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.2868. Min Act: -0.1700

Data Index: 1038291 (C4 (Web Text))

Max Activating Token Index: 85

Text #15

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.2711. Min Act: -0.1700

Data Index: 261990 (C4 (Web Text))

Max Activating Token Index: 465

Text #16

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3339. Min Act: -0.1700

Data Index: 607872 (C4 (Web Text))

Max Activating Token Index: 251

Text #17

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.2868. Min Act: -0.1700

Data Index: 134626 (C4 (Web Text))

Max Activating Token Index: 19

Text #18

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3182. Min Act: -0.1700

Data Index: 287709 (C4 (Web Text))

Max Activating Token Index: 705

Text #19

Max Range: 3.5847. Min Range: -3.5847

Max Act: 3.3496. Min Act: -0.1700

Data Index: 897003 (C4 (Web Text))

Max Activating Token Index: 758