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

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

Neuron 1913 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.2446. Min Act: -0.1700

Data Index: 1330315 (C4 (Web Text))

Max Activating Token Index: 1016

Text #1

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.2115. Min Act: -0.1700

Data Index: 770554 (C4 (Web Text))

Max Activating Token Index: 1015

Text #2

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1949. Min Act: -0.1700

Data Index: 1039600 (C4 (Web Text))

Max Activating Token Index: 1015

Text #3

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1949. Min Act: -0.1700

Data Index: 912974 (C4 (Web Text))

Max Activating Token Index: 1013

Text #4

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1949. Min Act: -0.1700

Data Index: 280979 (C4 (Web Text))

Max Activating Token Index: 1017

Text #5

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1866. Min Act: -0.1700

Data Index: 890066 (C4 (Web Text))

Max Activating Token Index: 1021

Text #6

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1949. Min Act: -0.1700

Data Index: 878965 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1866. Min Act: -0.1700

Data Index: 137910 (C4 (Web Text))

Max Activating Token Index: 1015

Text #8

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1783. Min Act: -0.1700

Data Index: 1163963 (C4 (Web Text))

Max Activating Token Index: 1014

Text #9

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1534. Min Act: -0.1700

Data Index: 1312692 (C4 (Web Text))

Max Activating Token Index: 1016

Text #10

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1700. Min Act: -0.1700

Data Index: 1148611 (C4 (Web Text))

Max Activating Token Index: 1011

Text #11

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1700. Min Act: -0.1700

Data Index: 4777 (C4 (Web Text))

Max Activating Token Index: 999

Text #12

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1700. Min Act: -0.1700

Data Index: 176661 (C4 (Web Text))

Max Activating Token Index: 1004

Text #13

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1617. Min Act: -0.1700

Data Index: 1063185 (C4 (Web Text))

Max Activating Token Index: 1016

Text #14

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1534. Min Act: -0.1700

Data Index: 1149688 (C4 (Web Text))

Max Activating Token Index: 1016

Text #15

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1617. Min Act: -0.1700

Data Index: 225130 (C4 (Web Text))

Max Activating Token Index: 1010

Text #16

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1534. Min Act: -0.1700

Data Index: 300824 (C4 (Web Text))

Max Activating Token Index: 1022

Text #17

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1367. Min Act: -0.1700

Data Index: 1273570 (C4 (Web Text))

Max Activating Token Index: 1017

Text #18

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1534. Min Act: -0.1700

Data Index: 1276287 (C4 (Web Text))

Max Activating Token Index: 1017

Text #19

Max Range: 2.2446. Min Range: -2.2446

Max Act: 2.1617. Min Act: -0.1700

Data Index: 1107709 (C4 (Web Text))

Max Activating Token Index: 1016