Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1439 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 6.0263. Min Range: -6.0263

Max Act: 6.0263. Min Act: -0.1699

Data Index: 440164 (C4 (Web Text))

Max Activating Token Index: 765

Text #1

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.8232. Min Act: -0.1699

Data Index: 882211 (C4 (Web Text))

Max Activating Token Index: 737

Text #2

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.6513. Min Act: -0.1699

Data Index: 1210110 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.6044. Min Act: -0.1699

Data Index: 899640 (C4 (Web Text))

Max Activating Token Index: 876

Text #4

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.4950. Min Act: -0.1699

Data Index: 566145 (C4 (Web Text))

Max Activating Token Index: 618

Text #5

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.4482. Min Act: -0.1699

Data Index: 1254772 (C4 (Web Text))

Max Activating Token Index: 909

Text #6

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.4169. Min Act: -0.1699

Data Index: 767095 (C4 (Web Text))

Max Activating Token Index: 906

Text #7

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.4325. Min Act: -0.1699

Data Index: 843814 (C4 (Web Text))

Max Activating Token Index: 580

Text #8

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.4169. Min Act: -0.1699

Data Index: 504514 (C4 (Web Text))

Max Activating Token Index: 177

Text #9

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2294. Min Act: -0.1699

Data Index: 421584 (C4 (Web Text))

Max Activating Token Index: 467

Text #10

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.3857. Min Act: -0.1699

Data Index: 392735 (C4 (Web Text))

Max Activating Token Index: 38

Text #11

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.3232. Min Act: -0.1699

Data Index: 915021 (C4 (Web Text))

Max Activating Token Index: 239

Text #12

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.3779. Min Act: -0.1699

Data Index: 784088 (C4 (Web Text))

Max Activating Token Index: 796

Text #13

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.3075. Min Act: -0.1699

Data Index: 1244457 (C4 (Web Text))

Max Activating Token Index: 222

Text #14

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2997. Min Act: -0.1699

Data Index: 218802 (C4 (Web Text))

Max Activating Token Index: 37

Text #15

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2294. Min Act: -0.1699

Data Index: 943547 (C4 (Web Text))

Max Activating Token Index: 271

Text #16

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.3075. Min Act: -0.1699

Data Index: 500243 (C4 (Web Text))

Max Activating Token Index: 508

Text #17

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2216. Min Act: -0.1699

Data Index: 1055945 (C4 (Web Text))

Max Activating Token Index: 966

Text #18

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2294. Min Act: -0.1699

Data Index: 1513306 (Python Code)

Max Activating Token Index: 218

Text #19

Max Range: 6.0263. Min Range: -6.0263

Max Act: 5.2216. Min Act: -0.1699

Data Index: 1047877 (C4 (Web Text))

Max Activating Token Index: 645