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

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

Neuron 1645 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: 3.5047. Min Range: -3.5047

Max Act: 3.5047. Min Act: -0.1700

Data Index: 351524 (C4 (Web Text))

Max Activating Token Index: 588

Text #1

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4891. Min Act: -0.1700

Data Index: 102744 (C4 (Web Text))

Max Activating Token Index: 88

Text #2

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4891. Min Act: -0.1700

Data Index: 303855 (C4 (Web Text))

Max Activating Token Index: 120

Text #3

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.5047. Min Act: -0.1700

Data Index: 627602 (C4 (Web Text))

Max Activating Token Index: 177

Text #4

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4577. Min Act: -0.1700

Data Index: 60162 (C4 (Web Text))

Max Activating Token Index: 705

Text #5

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4264. Min Act: -0.1700

Data Index: 1647738 (Python Code)

Max Activating Token Index: 304

Text #6

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4420. Min Act: -0.1700

Data Index: 156608 (C4 (Web Text))

Max Activating Token Index: 95

Text #7

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4420. Min Act: -0.1700

Data Index: 1304878 (C4 (Web Text))

Max Activating Token Index: 993

Text #8

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4264. Min Act: -0.1700

Data Index: 1158349 (C4 (Web Text))

Max Activating Token Index: 989

Text #9

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4420. Min Act: -0.1700

Data Index: 698339 (C4 (Web Text))

Max Activating Token Index: 709

Text #10

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4420. Min Act: -0.1700

Data Index: 667726 (C4 (Web Text))

Max Activating Token Index: 865

Text #11

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4264. Min Act: -0.1700

Data Index: 689852 (C4 (Web Text))

Max Activating Token Index: 579

Text #12

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.3950. Min Act: -0.1700

Data Index: 1269304 (C4 (Web Text))

Max Activating Token Index: 196

Text #13

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4107. Min Act: -0.1700

Data Index: 612355 (C4 (Web Text))

Max Activating Token Index: 559

Text #14

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.3950. Min Act: -0.1700

Data Index: 111500 (C4 (Web Text))

Max Activating Token Index: 352

Text #15

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4264. Min Act: -0.1700

Data Index: 313752 (C4 (Web Text))

Max Activating Token Index: 271

Text #16

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4107. Min Act: -0.1700

Data Index: 922319 (C4 (Web Text))

Max Activating Token Index: 464

Text #17

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4107. Min Act: -0.1700

Data Index: 303112 (C4 (Web Text))

Max Activating Token Index: 492

Text #18

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.4264. Min Act: -0.1700

Data Index: 753580 (C4 (Web Text))

Max Activating Token Index: 408

Text #19

Max Range: 3.5047. Min Range: -3.5047

Max Act: 3.3793. Min Act: -0.1700

Data Index: 841361 (C4 (Web Text))

Max Activating Token Index: 854