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

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

Neuron 1551 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: 9.5572. Min Range: -9.5572

Max Act: 9.5572. Min Act: -0.1700

Data Index: 74267 (C4 (Web Text))

Max Activating Token Index: 52

Text #1

Max Range: 9.5572. Min Range: -9.5572

Max Act: 9.0572. Min Act: -0.1700

Data Index: 174998 (C4 (Web Text))

Max Activating Token Index: 32

Text #2

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.9322. Min Act: -0.1700

Data Index: 334578 (C4 (Web Text))

Max Activating Token Index: 862

Text #3

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.8697. Min Act: -0.1700

Data Index: 436506 (C4 (Web Text))

Max Activating Token Index: 51

Text #4

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.9322. Min Act: -0.1700

Data Index: 164276 (C4 (Web Text))

Max Activating Token Index: 67

Text #5

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.6197. Min Act: -0.1700

Data Index: 1400865 (Python Code)

Max Activating Token Index: 797

Text #6

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.5572. Min Act: -0.1700

Data Index: 940133 (C4 (Web Text))

Max Activating Token Index: 58

Text #7

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.5572. Min Act: -0.1700

Data Index: 225257 (C4 (Web Text))

Max Activating Token Index: 237

Text #8

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.5572. Min Act: -0.1700

Data Index: 298387 (C4 (Web Text))

Max Activating Token Index: 9

Text #9

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.4322. Min Act: -0.1700

Data Index: 1293139 (C4 (Web Text))

Max Activating Token Index: 255

Text #10

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.4947. Min Act: -0.1700

Data Index: 255295 (C4 (Web Text))

Max Activating Token Index: 385

Text #11

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.5572. Min Act: -0.1700

Data Index: 373718 (C4 (Web Text))

Max Activating Token Index: 1007

Text #12

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.3697. Min Act: -0.1700

Data Index: 4147 (C4 (Web Text))

Max Activating Token Index: 704

Text #13

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.6197. Min Act: -0.1700

Data Index: 648244 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.3072. Min Act: -0.1700

Data Index: 411641 (C4 (Web Text))

Max Activating Token Index: 820

Text #15

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.3697. Min Act: -0.1700

Data Index: 650886 (C4 (Web Text))

Max Activating Token Index: 17

Text #16

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.1822. Min Act: -0.1700

Data Index: 474456 (C4 (Web Text))

Max Activating Token Index: 9

Text #17

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.2447. Min Act: -0.1700

Data Index: 513210 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.1822. Min Act: -0.1700

Data Index: 1025754 (C4 (Web Text))

Max Activating Token Index: 982

Text #19

Max Range: 9.5572. Min Range: -9.5572

Max Act: 8.0572. Min Act: -0.1700

Data Index: 597704 (C4 (Web Text))

Max Activating Token Index: 80