Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 3 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.5482. Min Act: -0.1700

Data Index: 1329960 (C4 (Web Text))

Max Activating Token Index: 542

Text #1

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.5012. Min Act: -0.1700

Data Index: 305094 (C4 (Web Text))

Max Activating Token Index: 72

Text #2

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4856. Min Act: -0.1700

Data Index: 965374 (C4 (Web Text))

Max Activating Token Index: 614

Text #3

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4699. Min Act: -0.1700

Data Index: 501882 (C4 (Web Text))

Max Activating Token Index: 355

Text #4

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4385. Min Act: -0.1700

Data Index: 448254 (C4 (Web Text))

Max Activating Token Index: 51

Text #5

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4072. Min Act: -0.1700

Data Index: 719105 (C4 (Web Text))

Max Activating Token Index: 55

Text #6

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4229. Min Act: -0.1700

Data Index: 1353774 (C4 (Web Text))

Max Activating Token Index: 867

Text #7

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4072. Min Act: -0.1700

Data Index: 103579 (C4 (Web Text))

Max Activating Token Index: 367

Text #8

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4229. Min Act: -0.1700

Data Index: 952160 (C4 (Web Text))

Max Activating Token Index: 88

Text #9

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4072. Min Act: -0.1700

Data Index: 1211482 (C4 (Web Text))

Max Activating Token Index: 136

Text #10

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4072. Min Act: -0.1700

Data Index: 476307 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3915. Min Act: -0.1700

Data Index: 35201 (C4 (Web Text))

Max Activating Token Index: 611

Text #12

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3601. Min Act: -0.1700

Data Index: 988123 (C4 (Web Text))

Max Activating Token Index: 746

Text #13

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.4072. Min Act: -0.1700

Data Index: 99176 (C4 (Web Text))

Max Activating Token Index: 482

Text #14

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3758. Min Act: -0.1700

Data Index: 685216 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3915. Min Act: -0.1700

Data Index: 1113403 (C4 (Web Text))

Max Activating Token Index: 497

Text #16

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3758. Min Act: -0.1700

Data Index: 143055 (C4 (Web Text))

Max Activating Token Index: 564

Text #17

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3601. Min Act: -0.1700

Data Index: 419706 (C4 (Web Text))

Max Activating Token Index: 818

Text #18

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3758. Min Act: -0.1700

Data Index: 862011 (C4 (Web Text))

Max Activating Token Index: 417

Text #19

Max Range: 3.5482. Min Range: -3.5482

Max Act: 3.3601. Min Act: -0.1700

Data Index: 732763 (C4 (Web Text))

Max Activating Token Index: 425