Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 519 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.8539. Min Act: -0.1700

Data Index: 190846 (C4 (Web Text))

Max Activating Token Index: 540

Text #1

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.7289. Min Act: -0.1700

Data Index: 47027 (C4 (Web Text))

Max Activating Token Index: 748

Text #2

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.6352. Min Act: -0.1699

Data Index: 738560 (C4 (Web Text))

Max Activating Token Index: 581

Text #3

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.5727. Min Act: -0.1700

Data Index: 816886 (C4 (Web Text))

Max Activating Token Index: 351

Text #4

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4789. Min Act: -0.1700

Data Index: 735558 (C4 (Web Text))

Max Activating Token Index: 143

Text #5

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4789. Min Act: -0.1700

Data Index: 3714 (C4 (Web Text))

Max Activating Token Index: 116

Text #6

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.5102. Min Act: -0.1700

Data Index: 309096 (C4 (Web Text))

Max Activating Token Index: 838

Text #7

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4789. Min Act: -0.1700

Data Index: 948427 (C4 (Web Text))

Max Activating Token Index: 141

Text #8

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4477. Min Act: -0.1700

Data Index: 728083 (C4 (Web Text))

Max Activating Token Index: 624

Text #9

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4477. Min Act: -0.1700

Data Index: 1010987 (C4 (Web Text))

Max Activating Token Index: 426

Text #10

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3227. Min Act: -0.1700

Data Index: 967963 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3852. Min Act: -0.1700

Data Index: 62858 (C4 (Web Text))

Max Activating Token Index: 344

Text #12

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4164. Min Act: -0.1700

Data Index: 574590 (C4 (Web Text))

Max Activating Token Index: 327

Text #13

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4164. Min Act: -0.1700

Data Index: 235258 (C4 (Web Text))

Max Activating Token Index: 110

Text #14

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3852. Min Act: -0.1699

Data Index: 198357 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3852. Min Act: -0.1699

Data Index: 338412 (C4 (Web Text))

Max Activating Token Index: 798

Text #16

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.4164. Min Act: -0.1700

Data Index: 290118 (C4 (Web Text))

Max Activating Token Index: 769

Text #17

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3539. Min Act: -0.1700

Data Index: 1094327 (C4 (Web Text))

Max Activating Token Index: 181

Text #18

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3852. Min Act: -0.1700

Data Index: 938883 (C4 (Web Text))

Max Activating Token Index: 457

Text #19

Max Range: 5.8539. Min Range: -5.8539

Max Act: 5.3539. Min Act: -0.1700

Data Index: 451792 (C4 (Web Text))

Max Activating Token Index: 698