Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 742 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.2574. Min Act: -0.1699

Data Index: 361688 (C4 (Web Text))

Max Activating Token Index: 830

Text #1

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.2262. Min Act: -0.1700

Data Index: 1110481 (C4 (Web Text))

Max Activating Token Index: 1017

Text #2

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.1324. Min Act: -0.1035

Data Index: 616790 (C4 (Web Text))

Max Activating Token Index: 217

Text #3

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.1012. Min Act: -0.1699

Data Index: 169134 (C4 (Web Text))

Max Activating Token Index: 43

Text #4

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.1324. Min Act: -0.1675

Data Index: 448746 (C4 (Web Text))

Max Activating Token Index: 120

Text #5

Max Range: 6.2574. Min Range: -6.2574

Max Act: 6.0387. Min Act: -0.1685

Data Index: 65347 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.9762. Min Act: -0.1566

Data Index: 62829 (C4 (Web Text))

Max Activating Token Index: 126

Text #7

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.9762. Min Act: -0.1699

Data Index: 203371 (C4 (Web Text))

Max Activating Token Index: 271

Text #8

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.9449. Min Act: -0.1700

Data Index: 719955 (C4 (Web Text))

Max Activating Token Index: 599

Text #9

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: -0.1610

Data Index: 997803 (C4 (Web Text))

Max Activating Token Index: 263

Text #10

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: -0.1698

Data Index: 874416 (C4 (Web Text))

Max Activating Token Index: 392

Text #11

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: -0.1618

Data Index: 805879 (C4 (Web Text))

Max Activating Token Index: 173

Text #12

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8512. Min Act: 0.1320

Data Index: 913453 (C4 (Web Text))

Max Activating Token Index: 78

Text #13

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: -0.1700

Data Index: 1616034 (Python Code)

Max Activating Token Index: 792

Text #14

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8199. Min Act: -0.1684

Data Index: 743102 (C4 (Web Text))

Max Activating Token Index: 631

Text #15

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8199. Min Act: 0.4293

Data Index: 541032 (C4 (Web Text))

Max Activating Token Index: 117

Text #16

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: 0.2637

Data Index: 709343 (C4 (Web Text))

Max Activating Token Index: 397

Text #17

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8199. Min Act: -0.1648

Data Index: 258197 (C4 (Web Text))

Max Activating Token Index: 62

Text #18

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8824. Min Act: -0.1332

Data Index: 711964 (C4 (Web Text))

Max Activating Token Index: 174

Text #19

Max Range: 6.2574. Min Range: -6.2574

Max Act: 5.8199. Min Act: -0.1681

Data Index: 1158749 (C4 (Web Text))

Max Activating Token Index: 892