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

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

Neuron 540 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: 3.8063. Min Range: -3.8063

Max Act: 3.8063. Min Act: -0.1700

Data Index: 4225 (C4 (Web Text))

Max Activating Token Index: 823

Text #1

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.6811. Min Act: -0.1700

Data Index: 11710 (C4 (Web Text))

Max Activating Token Index: 245

Text #2

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.6185. Min Act: -0.1700

Data Index: 1042337 (C4 (Web Text))

Max Activating Token Index: 908

Text #3

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.5559. Min Act: -0.1700

Data Index: 612171 (C4 (Web Text))

Max Activating Token Index: 496

Text #4

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.5245. Min Act: -0.1700

Data Index: 1254518 (C4 (Web Text))

Max Activating Token Index: 330

Text #5

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4932. Min Act: -0.1700

Data Index: 1274128 (C4 (Web Text))

Max Activating Token Index: 768

Text #6

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4775. Min Act: -0.1700

Data Index: 349698 (C4 (Web Text))

Max Activating Token Index: 461

Text #7

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4618. Min Act: -0.1700

Data Index: 682087 (C4 (Web Text))

Max Activating Token Index: 1018

Text #8

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4148. Min Act: -0.1700

Data Index: 347367 (C4 (Web Text))

Max Activating Token Index: 804

Text #9

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4462. Min Act: -0.1700

Data Index: 503499 (C4 (Web Text))

Max Activating Token Index: 985

Text #10

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4462. Min Act: -0.1700

Data Index: 1348266 (C4 (Web Text))

Max Activating Token Index: 587

Text #11

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4618. Min Act: -0.1700

Data Index: 136022 (C4 (Web Text))

Max Activating Token Index: 376

Text #12

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.3991. Min Act: -0.1700

Data Index: 1010929 (C4 (Web Text))

Max Activating Token Index: 538

Text #13

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.3834. Min Act: -0.1700

Data Index: 292576 (C4 (Web Text))

Max Activating Token Index: 298

Text #14

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4148. Min Act: -0.1700

Data Index: 1353764 (C4 (Web Text))

Max Activating Token Index: 175

Text #15

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4148. Min Act: -0.1700

Data Index: 521065 (C4 (Web Text))

Max Activating Token Index: 713

Text #16

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.4305. Min Act: -0.1700

Data Index: 102631 (C4 (Web Text))

Max Activating Token Index: 689

Text #17

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.3991. Min Act: -0.1700

Data Index: 195082 (C4 (Web Text))

Max Activating Token Index: 905

Text #18

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.3521. Min Act: -0.1700

Data Index: 655576 (C4 (Web Text))

Max Activating Token Index: 806

Text #19

Max Range: 3.8063. Min Range: -3.8063

Max Act: 3.3521. Min Act: -0.1700

Data Index: 1094068 (C4 (Web Text))

Max Activating Token Index: 944