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

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

Neuron 820 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.8022. Min Act: -0.1700

Data Index: 807389 (C4 (Web Text))

Max Activating Token Index: 806

Text #1

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.9272. Min Act: -0.1700

Data Index: 304033 (C4 (Web Text))

Max Activating Token Index: 922

Text #2

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.8647. Min Act: -0.1700

Data Index: 1143688 (C4 (Web Text))

Max Activating Token Index: 967

Text #3

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.4897. Min Act: -0.1699

Data Index: 724724 (C4 (Web Text))

Max Activating Token Index: 754

Text #4

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.3022. Min Act: -0.1700

Data Index: 523573 (C4 (Web Text))

Max Activating Token Index: 826

Text #5

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.1772. Min Act: -0.1700

Data Index: 206015 (C4 (Web Text))

Max Activating Token Index: 553

Text #6

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.1147. Min Act: -0.1700

Data Index: 610503 (C4 (Web Text))

Max Activating Token Index: 518

Text #7

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.8647. Min Act: -0.1700

Data Index: 351853 (C4 (Web Text))

Max Activating Token Index: 442

Text #8

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.0522. Min Act: -0.1700

Data Index: 1349812 (C4 (Web Text))

Max Activating Token Index: 666

Text #9

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.9897. Min Act: -0.1700

Data Index: 1168687 (C4 (Web Text))

Max Activating Token Index: 508

Text #10

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.8022. Min Act: -0.1698

Data Index: 600573 (C4 (Web Text))

Max Activating Token Index: 229

Text #11

Max Range: 7.9272. Min Range: -7.9272

Max Act: 7.2397. Min Act: -0.1700

Data Index: 1044002 (C4 (Web Text))

Max Activating Token Index: 796

Text #12

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.8022. Min Act: -0.1700

Data Index: 912324 (C4 (Web Text))

Max Activating Token Index: 224

Text #13

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.7397. Min Act: -0.1700

Data Index: 3279 (C4 (Web Text))

Max Activating Token Index: 110

Text #14

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.6147. Min Act: -0.1695

Data Index: 390516 (C4 (Web Text))

Max Activating Token Index: 54

Text #15

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.6147. Min Act: -0.1700

Data Index: 52193 (C4 (Web Text))

Max Activating Token Index: 245

Text #16

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.7397. Min Act: -0.1700

Data Index: 1134177 (C4 (Web Text))

Max Activating Token Index: 510

Text #17

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.5834. Min Act: -0.1700

Data Index: 658190 (C4 (Web Text))

Max Activating Token Index: 361

Text #18

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.5522. Min Act: -0.1700

Data Index: 20935 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 7.9272. Min Range: -7.9272

Max Act: 6.5209. Min Act: -0.1699

Data Index: 253673 (C4 (Web Text))

Max Activating Token Index: 157