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

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

Neuron 1041 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.4797. Min Act: -0.1700

Data Index: 301095 (C4 (Web Text))

Max Activating Token Index: 707

Text #1

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.4797. Min Act: -0.1700

Data Index: 1262055 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.4484. Min Act: -0.1700

Data Index: 787321 (C4 (Web Text))

Max Activating Token Index: 805

Text #3

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3857. Min Act: -0.1700

Data Index: 197330 (C4 (Web Text))

Max Activating Token Index: 233

Text #4

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3857. Min Act: -0.1700

Data Index: 277944 (C4 (Web Text))

Max Activating Token Index: 27

Text #5

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3700. Min Act: -0.1700

Data Index: 981766 (C4 (Web Text))

Max Activating Token Index: 689

Text #6

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3700. Min Act: -0.1700

Data Index: 276480 (C4 (Web Text))

Max Activating Token Index: 123

Text #7

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3229. Min Act: -0.1700

Data Index: 605195 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3229. Min Act: -0.1700

Data Index: 1332800 (C4 (Web Text))

Max Activating Token Index: 116

Text #9

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.3072. Min Act: -0.1700

Data Index: 155449 (C4 (Web Text))

Max Activating Token Index: 267

Text #10

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2758. Min Act: -0.1700

Data Index: 110738 (C4 (Web Text))

Max Activating Token Index: 847

Text #11

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2915. Min Act: -0.1700

Data Index: 627255 (C4 (Web Text))

Max Activating Token Index: 292

Text #12

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2443. Min Act: -0.1700

Data Index: 1244359 (C4 (Web Text))

Max Activating Token Index: 26

Text #13

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2600. Min Act: -0.1700

Data Index: 516039 (C4 (Web Text))

Max Activating Token Index: 152

Text #14

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2915. Min Act: -0.1700

Data Index: 780003 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2758. Min Act: -0.1700

Data Index: 1585014 (Python Code)

Max Activating Token Index: 14

Text #16

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2600. Min Act: -0.1700

Data Index: 73234 (C4 (Web Text))

Max Activating Token Index: 659

Text #17

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2443. Min Act: -0.1700

Data Index: 803563 (C4 (Web Text))

Max Activating Token Index: 962

Text #18

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2758. Min Act: -0.1700

Data Index: 1237846 (C4 (Web Text))

Max Activating Token Index: 986

Text #19

Max Range: 3.4797. Min Range: -3.4797

Max Act: 3.2758. Min Act: -0.1700

Data Index: 48978 (C4 (Web Text))

Max Activating Token Index: 259