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

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

Neuron 970 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.3412. Min Act: -0.1700

Data Index: 238090 (C4 (Web Text))

Max Activating Token Index: 40

Text #1

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.2313. Min Act: -0.1700

Data Index: 652648 (C4 (Web Text))

Max Activating Token Index: 785

Text #2

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.2313. Min Act: -0.1700

Data Index: 1131690 (C4 (Web Text))

Max Activating Token Index: 951

Text #3

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1998. Min Act: -0.1700

Data Index: 1497735 (Python Code)

Max Activating Token Index: 371

Text #4

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.2156. Min Act: -0.1700

Data Index: 1046439 (C4 (Web Text))

Max Activating Token Index: 383

Text #5

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.2156. Min Act: -0.1700

Data Index: 97783 (C4 (Web Text))

Max Activating Token Index: 782

Text #6

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1369. Min Act: -0.1700

Data Index: 415413 (C4 (Web Text))

Max Activating Token Index: 148

Text #7

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1526. Min Act: -0.1700

Data Index: 1083132 (C4 (Web Text))

Max Activating Token Index: 102

Text #8

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1683. Min Act: -0.1700

Data Index: 1073442 (C4 (Web Text))

Max Activating Token Index: 603

Text #9

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1683. Min Act: -0.1700

Data Index: 1526007 (Python Code)

Max Activating Token Index: 673

Text #10

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1526. Min Act: -0.1700

Data Index: 1045107 (C4 (Web Text))

Max Activating Token Index: 263

Text #11

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1369. Min Act: -0.1700

Data Index: 1387300 (Python Code)

Max Activating Token Index: 401

Text #12

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1369. Min Act: -0.1700

Data Index: 1287124 (C4 (Web Text))

Max Activating Token Index: 817

Text #13

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1211. Min Act: -0.1700

Data Index: 851659 (C4 (Web Text))

Max Activating Token Index: 438

Text #14

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1211. Min Act: -0.1700

Data Index: 443254 (C4 (Web Text))

Max Activating Token Index: 134

Text #15

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.1053. Min Act: -0.1700

Data Index: 317851 (C4 (Web Text))

Max Activating Token Index: 312

Text #16

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.0738. Min Act: -0.1700

Data Index: 1017577 (C4 (Web Text))

Max Activating Token Index: 613

Text #17

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.0580. Min Act: -0.1700

Data Index: 136071 (C4 (Web Text))

Max Activating Token Index: 932

Text #18

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.0738. Min Act: -0.1700

Data Index: 961900 (C4 (Web Text))

Max Activating Token Index: 827

Text #19

Max Range: 3.3412. Min Range: -3.3412

Max Act: 3.0738. Min Act: -0.1700

Data Index: 73546 (C4 (Web Text))

Max Activating Token Index: 79