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

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

Neuron 275 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: 3.2757. Min Range: -3.2757

Max Act: 3.2757. Min Act: -0.1700

Data Index: 852774 (C4 (Web Text))

Max Activating Token Index: 658

Text #1

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.2286. Min Act: -0.1700

Data Index: 916742 (C4 (Web Text))

Max Activating Token Index: 518

Text #2

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.2128. Min Act: -0.1700

Data Index: 211627 (C4 (Web Text))

Max Activating Token Index: 480

Text #3

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.1184. Min Act: -0.1700

Data Index: 337294 (C4 (Web Text))

Max Activating Token Index: 469

Text #4

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0868. Min Act: -0.1700

Data Index: 42930 (C4 (Web Text))

Max Activating Token Index: 482

Text #5

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0553. Min Act: -0.1700

Data Index: 1115398 (C4 (Web Text))

Max Activating Token Index: 816

Text #6

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0868. Min Act: -0.1700

Data Index: 647695 (C4 (Web Text))

Max Activating Token Index: 529

Text #7

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0711. Min Act: -0.1700

Data Index: 770930 (C4 (Web Text))

Max Activating Token Index: 375

Text #8

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 139308 (C4 (Web Text))

Max Activating Token Index: 768

Text #9

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0237. Min Act: -0.1700

Data Index: 606823 (C4 (Web Text))

Max Activating Token Index: 658

Text #10

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 825797 (C4 (Web Text))

Max Activating Token Index: 1003

Text #11

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0395. Min Act: -0.1700

Data Index: 1229662 (C4 (Web Text))

Max Activating Token Index: 942

Text #12

Max Range: 3.2757. Min Range: -3.2757

Max Act: 2.9446. Min Act: -0.1700

Data Index: 1225511 (C4 (Web Text))

Max Activating Token Index: 899

Text #13

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 475725 (C4 (Web Text))

Max Activating Token Index: 232

Text #14

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 471703 (C4 (Web Text))

Max Activating Token Index: 635

Text #15

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 38030 (C4 (Web Text))

Max Activating Token Index: 585

Text #16

Max Range: 3.2757. Min Range: -3.2757

Max Act: 2.9921. Min Act: -0.1700

Data Index: 849681 (C4 (Web Text))

Max Activating Token Index: 746

Text #17

Max Range: 3.2757. Min Range: -3.2757

Max Act: 3.0079. Min Act: -0.1700

Data Index: 885498 (C4 (Web Text))

Max Activating Token Index: 330

Text #18

Max Range: 3.2757. Min Range: -3.2757

Max Act: 2.9604. Min Act: -0.1700

Data Index: 538454 (C4 (Web Text))

Max Activating Token Index: 455

Text #19

Max Range: 3.2757. Min Range: -3.2757

Max Act: 2.9604. Min Act: -0.1700

Data Index: 1470035 (Python Code)

Max Activating Token Index: 26