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

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

Neuron 820 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3767. Min Act: -0.1700

Data Index: 1248127 (C4 (Web Text))

Max Activating Token Index: 613

Text #1

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3924. Min Act: -0.1700

Data Index: 1304297 (C4 (Web Text))

Max Activating Token Index: 447

Text #2

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3767. Min Act: -0.1700

Data Index: 1340886 (C4 (Web Text))

Max Activating Token Index: 7

Text #3

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3611. Min Act: -0.1700

Data Index: 1059261 (C4 (Web Text))

Max Activating Token Index: 661

Text #4

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3767. Min Act: -0.1700

Data Index: 807968 (C4 (Web Text))

Max Activating Token Index: 48

Text #5

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3611. Min Act: -0.1700

Data Index: 155297 (C4 (Web Text))

Max Activating Token Index: 921

Text #6

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3454. Min Act: -0.1700

Data Index: 1070764 (C4 (Web Text))

Max Activating Token Index: 25

Text #7

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3454. Min Act: -0.1700

Data Index: 11641 (C4 (Web Text))

Max Activating Token Index: 220

Text #8

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3454. Min Act: -0.1700

Data Index: 865299 (C4 (Web Text))

Max Activating Token Index: 90

Text #9

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3297. Min Act: -0.1700

Data Index: 444157 (C4 (Web Text))

Max Activating Token Index: 435

Text #10

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3454. Min Act: -0.1700

Data Index: 11814 (C4 (Web Text))

Max Activating Token Index: 518

Text #11

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3297. Min Act: -0.1700

Data Index: 671162 (C4 (Web Text))

Max Activating Token Index: 210

Text #12

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3297. Min Act: -0.1700

Data Index: 335331 (C4 (Web Text))

Max Activating Token Index: 17

Text #13

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3454. Min Act: -0.1700

Data Index: 241013 (C4 (Web Text))

Max Activating Token Index: 670

Text #14

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3297. Min Act: -0.1700

Data Index: 672314 (C4 (Web Text))

Max Activating Token Index: 82

Text #15

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3140. Min Act: -0.1700

Data Index: 1084860 (C4 (Web Text))

Max Activating Token Index: 717

Text #16

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.3140. Min Act: -0.1700

Data Index: 1202875 (C4 (Web Text))

Max Activating Token Index: 8

Text #17

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.2983. Min Act: -0.1700

Data Index: 1148002 (C4 (Web Text))

Max Activating Token Index: 925

Text #18

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.2983. Min Act: -0.1700

Data Index: 411095 (C4 (Web Text))

Max Activating Token Index: 983

Text #19

Max Range: 3.3924. Min Range: -3.3924

Max Act: 3.2983. Min Act: -0.1700

Data Index: 660757 (C4 (Web Text))

Max Activating Token Index: 96