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

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

Neuron 785 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: 3.7557. Min Range: -3.7557

Max Act: 3.7557. Min Act: -0.1700

Data Index: 1648237 (Python Code)

Max Activating Token Index: 393

Text #1

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.5523. Min Act: -0.1700

Data Index: 694598 (C4 (Web Text))

Max Activating Token Index: 905

Text #2

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.4426. Min Act: -0.1700

Data Index: 1228907 (C4 (Web Text))

Max Activating Token Index: 683

Text #3

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3955. Min Act: -0.1700

Data Index: 1562650 (Python Code)

Max Activating Token Index: 313

Text #4

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.4269. Min Act: -0.1700

Data Index: 1605662 (Python Code)

Max Activating Token Index: 897

Text #5

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3642. Min Act: -0.1700

Data Index: 1048365 (C4 (Web Text))

Max Activating Token Index: 773

Text #6

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3955. Min Act: -0.1700

Data Index: 1158794 (C4 (Web Text))

Max Activating Token Index: 780

Text #7

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3955. Min Act: -0.1700

Data Index: 800703 (C4 (Web Text))

Max Activating Token Index: 729

Text #8

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3955. Min Act: -0.1700

Data Index: 809713 (C4 (Web Text))

Max Activating Token Index: 247

Text #9

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3642. Min Act: -0.1700

Data Index: 839153 (C4 (Web Text))

Max Activating Token Index: 78

Text #10

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3328. Min Act: -0.1700

Data Index: 23498 (C4 (Web Text))

Max Activating Token Index: 517

Text #11

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3328. Min Act: -0.1700

Data Index: 643080 (C4 (Web Text))

Max Activating Token Index: 748

Text #12

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3171. Min Act: -0.1700

Data Index: 1173240 (C4 (Web Text))

Max Activating Token Index: 65

Text #13

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3171. Min Act: -0.1700

Data Index: 1335034 (C4 (Web Text))

Max Activating Token Index: 681

Text #14

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3171. Min Act: -0.1700

Data Index: 196603 (C4 (Web Text))

Max Activating Token Index: 240

Text #15

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3328. Min Act: -0.1700

Data Index: 16098 (C4 (Web Text))

Max Activating Token Index: 667

Text #16

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3328. Min Act: -0.1700

Data Index: 638344 (C4 (Web Text))

Max Activating Token Index: 949

Text #17

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.3014. Min Act: -0.1700

Data Index: 239042 (C4 (Web Text))

Max Activating Token Index: 880

Text #18

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.2857. Min Act: -0.1700

Data Index: 248067 (C4 (Web Text))

Max Activating Token Index: 816

Text #19

Max Range: 3.7557. Min Range: -3.7557

Max Act: 3.2699. Min Act: -0.1700

Data Index: 24809 (C4 (Web Text))

Max Activating Token Index: 610