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

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

Neuron 1057 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: 3.2405. Min Range: -3.2405

Max Act: 3.2405. Min Act: -0.1700

Data Index: 67742 (C4 (Web Text))

Max Activating Token Index: 794

Text #1

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.2091. Min Act: -0.1700

Data Index: 33 (C4 (Web Text))

Max Activating Token Index: 551

Text #2

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.1461. Min Act: -0.1700

Data Index: 640648 (C4 (Web Text))

Max Activating Token Index: 290

Text #3

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.1461. Min Act: -0.1700

Data Index: 532777 (C4 (Web Text))

Max Activating Token Index: 891

Text #4

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0831. Min Act: -0.1700

Data Index: 1248263 (C4 (Web Text))

Max Activating Token Index: 123

Text #5

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0831. Min Act: -0.1700

Data Index: 90825 (C4 (Web Text))

Max Activating Token Index: 784

Text #6

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0673. Min Act: -0.1700

Data Index: 1229008 (C4 (Web Text))

Max Activating Token Index: 942

Text #7

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0515. Min Act: -0.1700

Data Index: 1319220 (C4 (Web Text))

Max Activating Token Index: 756

Text #8

Max Range: 3.2405. Min Range: -3.2405

Max Act: 2.9883. Min Act: -0.1700

Data Index: 516617 (C4 (Web Text))

Max Activating Token Index: 801

Text #9

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0515. Min Act: -0.1700

Data Index: 731053 (C4 (Web Text))

Max Activating Token Index: 985

Text #10

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0515. Min Act: -0.1700

Data Index: 792591 (C4 (Web Text))

Max Activating Token Index: 755

Text #11

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0515. Min Act: -0.1700

Data Index: 691195 (C4 (Web Text))

Max Activating Token Index: 89

Text #12

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0199. Min Act: -0.1700

Data Index: 343786 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0357. Min Act: -0.1700

Data Index: 164387 (C4 (Web Text))

Max Activating Token Index: 86

Text #14

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0357. Min Act: -0.1700

Data Index: 939887 (C4 (Web Text))

Max Activating Token Index: 737

Text #15

Max Range: 3.2405. Min Range: -3.2405

Max Act: 3.0357. Min Act: -0.1700

Data Index: 482718 (C4 (Web Text))

Max Activating Token Index: 728

Text #16

Max Range: 3.2405. Min Range: -3.2405

Max Act: 2.9725. Min Act: -0.1700

Data Index: 197850 (C4 (Web Text))

Max Activating Token Index: 679

Text #17

Max Range: 3.2405. Min Range: -3.2405

Max Act: 2.9883. Min Act: -0.1700

Data Index: 51152 (C4 (Web Text))

Max Activating Token Index: 859

Text #18

Max Range: 3.2405. Min Range: -3.2405

Max Act: 2.9883. Min Act: -0.1700

Data Index: 954832 (C4 (Web Text))

Max Activating Token Index: 670

Text #19

Max Range: 3.2405. Min Range: -3.2405

Max Act: 2.9883. Min Act: -0.1700

Data Index: 837745 (C4 (Web Text))

Max Activating Token Index: 217