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

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

Neuron 250 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.5584. Min Act: -0.1700

Data Index: 1098880 (C4 (Web Text))

Max Activating Token Index: 830

Text #1

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.4334. Min Act: -0.1700

Data Index: 981654 (C4 (Web Text))

Max Activating Token Index: 662

Text #2

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.4334. Min Act: -0.1700

Data Index: 909238 (C4 (Web Text))

Max Activating Token Index: 221

Text #3

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.3396. Min Act: -0.1700

Data Index: 1205420 (C4 (Web Text))

Max Activating Token Index: 155

Text #4

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.2771. Min Act: -0.1700

Data Index: 1015978 (C4 (Web Text))

Max Activating Token Index: 906

Text #5

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.2146. Min Act: -0.1700

Data Index: 739922 (C4 (Web Text))

Max Activating Token Index: 78

Text #6

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.2146. Min Act: -0.1700

Data Index: 322992 (C4 (Web Text))

Max Activating Token Index: 147

Text #7

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.2459. Min Act: -0.1700

Data Index: 1027650 (C4 (Web Text))

Max Activating Token Index: 915

Text #8

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1521. Min Act: -0.1700

Data Index: 1194416 (C4 (Web Text))

Max Activating Token Index: 1021

Text #9

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1834. Min Act: -0.1700

Data Index: 1186188 (C4 (Web Text))

Max Activating Token Index: 864

Text #10

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1521. Min Act: -0.1700

Data Index: 860615 (C4 (Web Text))

Max Activating Token Index: 939

Text #11

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1209. Min Act: -0.1700

Data Index: 1274604 (C4 (Web Text))

Max Activating Token Index: 932

Text #12

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1209. Min Act: -0.1700

Data Index: 787956 (C4 (Web Text))

Max Activating Token Index: 864

Text #13

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1521. Min Act: -0.1700

Data Index: 1190463 (C4 (Web Text))

Max Activating Token Index: 902

Text #14

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1834. Min Act: -0.1700

Data Index: 554954 (C4 (Web Text))

Max Activating Token Index: 723

Text #15

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1209. Min Act: -0.1700

Data Index: 1301950 (C4 (Web Text))

Max Activating Token Index: 655

Text #16

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1209. Min Act: -0.1700

Data Index: 541918 (C4 (Web Text))

Max Activating Token Index: 410

Text #17

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.1521. Min Act: -0.1700

Data Index: 758356 (C4 (Web Text))

Max Activating Token Index: 746

Text #18

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.0584. Min Act: -0.1700

Data Index: 707042 (C4 (Web Text))

Max Activating Token Index: 249

Text #19

Max Range: 5.5584. Min Range: -5.5584

Max Act: 5.0271. Min Act: -0.1700

Data Index: 975420 (C4 (Web Text))

Max Activating Token Index: 852