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

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

Neuron 67 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: 3.8567. Min Range: -3.8567

Max Act: 3.8567. Min Act: -0.1700

Data Index: 1504365 (Python Code)

Max Activating Token Index: 978

Text #1

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8254. Min Act: -0.1700

Data Index: 864242 (C4 (Web Text))

Max Activating Token Index: 596

Text #2

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7942. Min Act: -0.1700

Data Index: 1530995 (Python Code)

Max Activating Token Index: 460

Text #3

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8567. Min Act: -0.1700

Data Index: 770945 (C4 (Web Text))

Max Activating Token Index: 533

Text #4

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8567. Min Act: -0.1700

Data Index: 733205 (C4 (Web Text))

Max Activating Token Index: 421

Text #5

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8098. Min Act: -0.1700

Data Index: 489413 (C4 (Web Text))

Max Activating Token Index: 69

Text #6

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8254. Min Act: -0.1700

Data Index: 1440848 (Python Code)

Max Activating Token Index: 765

Text #7

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.8098. Min Act: -0.1700

Data Index: 1201121 (C4 (Web Text))

Max Activating Token Index: 78

Text #8

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7629. Min Act: -0.1700

Data Index: 246972 (C4 (Web Text))

Max Activating Token Index: 359

Text #9

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7629. Min Act: -0.1700

Data Index: 1458142 (Python Code)

Max Activating Token Index: 849

Text #10

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7472. Min Act: -0.1700

Data Index: 441414 (C4 (Web Text))

Max Activating Token Index: 303

Text #11

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7472. Min Act: -0.1700

Data Index: 1482550 (Python Code)

Max Activating Token Index: 962

Text #12

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7159. Min Act: -0.1700

Data Index: 189134 (C4 (Web Text))

Max Activating Token Index: 925

Text #13

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7316. Min Act: -0.1700

Data Index: 978756 (C4 (Web Text))

Max Activating Token Index: 863

Text #14

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7159. Min Act: -0.1700

Data Index: 280838 (C4 (Web Text))

Max Activating Token Index: 725

Text #15

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1199776 (C4 (Web Text))

Max Activating Token Index: 972

Text #16

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7003. Min Act: -0.1700

Data Index: 1202696 (C4 (Web Text))

Max Activating Token Index: 485

Text #17

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7159. Min Act: -0.1700

Data Index: 505436 (C4 (Web Text))

Max Activating Token Index: 376

Text #18

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.7003. Min Act: -0.1700

Data Index: 714085 (C4 (Web Text))

Max Activating Token Index: 971

Text #19

Max Range: 3.8567. Min Range: -3.8567

Max Act: 3.6846. Min Act: -0.1700

Data Index: 392490 (C4 (Web Text))

Max Activating Token Index: 1021