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

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

Neuron 1256 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.2428. Min Act: -0.1700

Data Index: 113134 (C4 (Web Text))

Max Activating Token Index: 521

Text #1

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.2428. Min Act: -0.1700

Data Index: 1055479 (C4 (Web Text))

Max Activating Token Index: 353

Text #2

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.0866. Min Act: -0.1700

Data Index: 1274961 (C4 (Web Text))

Max Activating Token Index: 19

Text #3

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.1178. Min Act: -0.1700

Data Index: 124431 (C4 (Web Text))

Max Activating Token Index: 388

Text #4

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.1491. Min Act: -0.1700

Data Index: 856508 (C4 (Web Text))

Max Activating Token Index: 136

Text #5

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9928. Min Act: -0.1700

Data Index: 1002975 (C4 (Web Text))

Max Activating Token Index: 883

Text #6

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.0553. Min Act: -0.1700

Data Index: 905245 (C4 (Web Text))

Max Activating Token Index: 191

Text #7

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.0553. Min Act: -0.1700

Data Index: 1148106 (C4 (Web Text))

Max Activating Token Index: 939

Text #8

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.0553. Min Act: -0.1700

Data Index: 847818 (C4 (Web Text))

Max Activating Token Index: 355

Text #9

Max Range: 5.2428. Min Range: -5.2428

Max Act: 5.0553. Min Act: -0.1700

Data Index: 344314 (C4 (Web Text))

Max Activating Token Index: 804

Text #10

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9928. Min Act: -0.1700

Data Index: 305574 (C4 (Web Text))

Max Activating Token Index: 89

Text #11

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9928. Min Act: -0.1700

Data Index: 288105 (C4 (Web Text))

Max Activating Token Index: 669

Text #12

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9616. Min Act: -0.1700

Data Index: 376460 (C4 (Web Text))

Max Activating Token Index: 885

Text #13

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9303. Min Act: -0.1700

Data Index: 1325650 (C4 (Web Text))

Max Activating Token Index: 209

Text #14

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9303. Min Act: -0.1698

Data Index: 646996 (C4 (Web Text))

Max Activating Token Index: 136

Text #15

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.8991. Min Act: -0.1700

Data Index: 383485 (C4 (Web Text))

Max Activating Token Index: 947

Text #16

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9303. Min Act: -0.1699

Data Index: 221865 (C4 (Web Text))

Max Activating Token Index: 290

Text #17

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9616. Min Act: -0.1700

Data Index: 136962 (C4 (Web Text))

Max Activating Token Index: 1012

Text #18

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.9303. Min Act: -0.1699

Data Index: 1162104 (C4 (Web Text))

Max Activating Token Index: 182

Text #19

Max Range: 5.2428. Min Range: -5.2428

Max Act: 4.8991. Min Act: -0.1700

Data Index: 434806 (C4 (Web Text))

Max Activating Token Index: 612