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

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

Neuron 688 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.6987. Min Act: -0.1700

Data Index: 514548 (C4 (Web Text))

Max Activating Token Index: 559

Text #1

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.6048. Min Act: -0.1700

Data Index: 1244728 (C4 (Web Text))

Max Activating Token Index: 556

Text #2

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.5735. Min Act: -0.1700

Data Index: 514830 (C4 (Web Text))

Max Activating Token Index: 441

Text #3

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.5108. Min Act: -0.1700

Data Index: 1161421 (C4 (Web Text))

Max Activating Token Index: 997

Text #4

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.4481. Min Act: -0.1700

Data Index: 1348979 (C4 (Web Text))

Max Activating Token Index: 236

Text #5

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.4168. Min Act: -0.1700

Data Index: 1335930 (C4 (Web Text))

Max Activating Token Index: 284

Text #6

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.4481. Min Act: -0.1700

Data Index: 502006 (C4 (Web Text))

Max Activating Token Index: 704

Text #7

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.4168. Min Act: -0.1700

Data Index: 901495 (C4 (Web Text))

Max Activating Token Index: 225

Text #8

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3226. Min Act: -0.1700

Data Index: 671903 (C4 (Web Text))

Max Activating Token Index: 918

Text #9

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3226. Min Act: -0.1700

Data Index: 845276 (C4 (Web Text))

Max Activating Token Index: 16

Text #10

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3540. Min Act: -0.1700

Data Index: 1055146 (C4 (Web Text))

Max Activating Token Index: 385

Text #11

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3226. Min Act: -0.1700

Data Index: 402158 (C4 (Web Text))

Max Activating Token Index: 928

Text #12

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3069. Min Act: -0.1700

Data Index: 1314678 (C4 (Web Text))

Max Activating Token Index: 25

Text #13

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3069. Min Act: -0.1700

Data Index: 1213591 (C4 (Web Text))

Max Activating Token Index: 19

Text #14

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3069. Min Act: -0.1700

Data Index: 670439 (C4 (Web Text))

Max Activating Token Index: 213

Text #15

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2755. Min Act: -0.1700

Data Index: 1588538 (Python Code)

Max Activating Token Index: 908

Text #16

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2126. Min Act: -0.1700

Data Index: 709882 (C4 (Web Text))

Max Activating Token Index: 539

Text #17

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2755. Min Act: -0.1700

Data Index: 864549 (C4 (Web Text))

Max Activating Token Index: 805

Text #18

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2598. Min Act: -0.1700

Data Index: 1009234 (C4 (Web Text))

Max Activating Token Index: 642

Text #19

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2126. Min Act: -0.1700

Data Index: 440633 (C4 (Web Text))

Max Activating Token Index: 405