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

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

Neuron 1038 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.3827. Min Act: -0.1700

Data Index: 669715 (C4 (Web Text))

Max Activating Token Index: 698

Text #1

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1952. Min Act: -0.1700

Data Index: 1122910 (C4 (Web Text))

Max Activating Token Index: 570

Text #2

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.2265. Min Act: -0.1700

Data Index: 15833 (C4 (Web Text))

Max Activating Token Index: 730

Text #3

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1640. Min Act: -0.1700

Data Index: 832964 (C4 (Web Text))

Max Activating Token Index: 600

Text #4

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.2265. Min Act: -0.1700

Data Index: 311603 (C4 (Web Text))

Max Activating Token Index: 662

Text #5

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1952. Min Act: -0.1700

Data Index: 780862 (C4 (Web Text))

Max Activating Token Index: 548

Text #6

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.2265. Min Act: -0.1700

Data Index: 749049 (C4 (Web Text))

Max Activating Token Index: 616

Text #7

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1952. Min Act: -0.1700

Data Index: 595004 (C4 (Web Text))

Max Activating Token Index: 611

Text #8

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1327. Min Act: -0.1700

Data Index: 1359096 (C4 (Web Text))

Max Activating Token Index: 485

Text #9

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1640. Min Act: -0.1700

Data Index: 239883 (C4 (Web Text))

Max Activating Token Index: 283

Text #10

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1952. Min Act: -0.1700

Data Index: 203332 (C4 (Web Text))

Max Activating Token Index: 223

Text #11

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1952. Min Act: -0.1700

Data Index: 374934 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1640. Min Act: -0.1700

Data Index: 376179 (C4 (Web Text))

Max Activating Token Index: 665

Text #13

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1015. Min Act: -0.1700

Data Index: 990206 (C4 (Web Text))

Max Activating Token Index: 445

Text #14

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1327. Min Act: -0.1700

Data Index: 598650 (C4 (Web Text))

Max Activating Token Index: 534

Text #15

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1640. Min Act: -0.1700

Data Index: 883745 (C4 (Web Text))

Max Activating Token Index: 182

Text #16

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1327. Min Act: -0.1700

Data Index: 1217481 (C4 (Web Text))

Max Activating Token Index: 549

Text #17

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.1327. Min Act: -0.1700

Data Index: 475257 (C4 (Web Text))

Max Activating Token Index: 178

Text #18

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.0702. Min Act: -0.1700

Data Index: 774330 (C4 (Web Text))

Max Activating Token Index: 168

Text #19

Max Range: 5.3827. Min Range: -5.3827

Max Act: 5.0390. Min Act: -0.1700

Data Index: 158614 (C4 (Web Text))

Max Activating Token Index: 449