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

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

Neuron 637 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.7477. Min Act: -0.1700

Data Index: 1201017 (C4 (Web Text))

Max Activating Token Index: 367

Text #1

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.7477. Min Act: -0.1700

Data Index: 1249657 (C4 (Web Text))

Max Activating Token Index: 685

Text #2

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6694. Min Act: -0.1700

Data Index: 1590387 (Python Code)

Max Activating Token Index: 210

Text #3

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6694. Min Act: -0.1700

Data Index: 320866 (C4 (Web Text))

Max Activating Token Index: 162

Text #4

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6694. Min Act: -0.1700

Data Index: 411817 (C4 (Web Text))

Max Activating Token Index: 351

Text #5

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6225. Min Act: -0.1700

Data Index: 409158 (C4 (Web Text))

Max Activating Token Index: 234

Text #6

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6538. Min Act: -0.1700

Data Index: 1109600 (C4 (Web Text))

Max Activating Token Index: 995

Text #7

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6225. Min Act: -0.1700

Data Index: 1106550 (C4 (Web Text))

Max Activating Token Index: 542

Text #8

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6381. Min Act: -0.1700

Data Index: 1435144 (Python Code)

Max Activating Token Index: 25

Text #9

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6068. Min Act: -0.1700

Data Index: 1397091 (Python Code)

Max Activating Token Index: 442

Text #10

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6068. Min Act: -0.1700

Data Index: 1623256 (Python Code)

Max Activating Token Index: 50

Text #11

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5912. Min Act: -0.1700

Data Index: 1475135 (Python Code)

Max Activating Token Index: 409

Text #12

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5912. Min Act: -0.1700

Data Index: 1494919 (Python Code)

Max Activating Token Index: 26

Text #13

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.6225. Min Act: -0.1700

Data Index: 16618 (C4 (Web Text))

Max Activating Token Index: 967

Text #14

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5912. Min Act: -0.1700

Data Index: 1095598 (C4 (Web Text))

Max Activating Token Index: 10

Text #15

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5755. Min Act: -0.1700

Data Index: 1544756 (Python Code)

Max Activating Token Index: 546

Text #16

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5755. Min Act: -0.1700

Data Index: 31800 (C4 (Web Text))

Max Activating Token Index: 287

Text #17

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5755. Min Act: -0.1700

Data Index: 355266 (C4 (Web Text))

Max Activating Token Index: 230

Text #18

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5755. Min Act: -0.1700

Data Index: 1428229 (Python Code)

Max Activating Token Index: 419

Text #19

Max Range: 3.7477. Min Range: -3.7477

Max Act: 3.5755. Min Act: -0.1700

Data Index: 1420572 (Python Code)

Max Activating Token Index: 405