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

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

Neuron 1415 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: 5.7258. Min Range: -5.7258

Max Act: 5.7258. Min Act: -0.1700

Data Index: 690123 (C4 (Web Text))

Max Activating Token Index: 629

Text #1

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.6946. Min Act: -0.1700

Data Index: 1312310 (C4 (Web Text))

Max Activating Token Index: 650

Text #2

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.6633. Min Act: -0.1700

Data Index: 1025896 (C4 (Web Text))

Max Activating Token Index: 445

Text #3

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.5071. Min Act: -0.1700

Data Index: 500326 (C4 (Web Text))

Max Activating Token Index: 254

Text #4

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.5071. Min Act: -0.1700

Data Index: 1244555 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4758. Min Act: -0.1700

Data Index: 1104194 (C4 (Web Text))

Max Activating Token Index: 31

Text #6

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4758. Min Act: -0.1700

Data Index: 815583 (C4 (Web Text))

Max Activating Token Index: 435

Text #7

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4446. Min Act: -0.1700

Data Index: 1145976 (C4 (Web Text))

Max Activating Token Index: 436

Text #8

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4758. Min Act: -0.1699

Data Index: 389775 (C4 (Web Text))

Max Activating Token Index: 719

Text #9

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4446. Min Act: -0.1700

Data Index: 97223 (C4 (Web Text))

Max Activating Token Index: 737

Text #10

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4758. Min Act: -0.1700

Data Index: 597089 (C4 (Web Text))

Max Activating Token Index: 648

Text #11

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4446. Min Act: -0.1700

Data Index: 434545 (C4 (Web Text))

Max Activating Token Index: 926

Text #12

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4133. Min Act: -0.1699

Data Index: 36960 (C4 (Web Text))

Max Activating Token Index: 687

Text #13

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.3821. Min Act: -0.1700

Data Index: 348998 (C4 (Web Text))

Max Activating Token Index: 716

Text #14

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4446. Min Act: -0.1700

Data Index: 361289 (C4 (Web Text))

Max Activating Token Index: 912

Text #15

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.3508. Min Act: -0.1700

Data Index: 359688 (C4 (Web Text))

Max Activating Token Index: 181

Text #16

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.3821. Min Act: -0.1700

Data Index: 493105 (C4 (Web Text))

Max Activating Token Index: 262

Text #17

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.4133. Min Act: -0.1700

Data Index: 81688 (C4 (Web Text))

Max Activating Token Index: 311

Text #18

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.3196. Min Act: -0.1700

Data Index: 1076012 (C4 (Web Text))

Max Activating Token Index: 520

Text #19

Max Range: 5.7258. Min Range: -5.7258

Max Act: 5.3508. Min Act: -0.1700

Data Index: 1003218 (C4 (Web Text))

Max Activating Token Index: 944