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

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

Neuron 2022 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: 3.8048. Min Range: -3.8048

Max Act: 3.8048. Min Act: -0.1700

Data Index: 131985 (C4 (Web Text))

Max Activating Token Index: 955

Text #1

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.6797. Min Act: -0.1700

Data Index: 1621341 (Python Code)

Max Activating Token Index: 812

Text #2

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.5857. Min Act: -0.1700

Data Index: 72371 (C4 (Web Text))

Max Activating Token Index: 898

Text #3

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.5857. Min Act: -0.1700

Data Index: 1131652 (C4 (Web Text))

Max Activating Token Index: 864

Text #4

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.5544. Min Act: -0.1700

Data Index: 809661 (C4 (Web Text))

Max Activating Token Index: 1004

Text #5

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4604. Min Act: -0.1700

Data Index: 567018 (C4 (Web Text))

Max Activating Token Index: 923

Text #6

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4604. Min Act: -0.1700

Data Index: 973904 (C4 (Web Text))

Max Activating Token Index: 951

Text #7

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4917. Min Act: -0.1700

Data Index: 918921 (C4 (Web Text))

Max Activating Token Index: 902

Text #8

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4917. Min Act: -0.1700

Data Index: 1177194 (C4 (Web Text))

Max Activating Token Index: 874

Text #9

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4917. Min Act: -0.1700

Data Index: 624726 (C4 (Web Text))

Max Activating Token Index: 917

Text #10

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.3977. Min Act: -0.1700

Data Index: 124174 (C4 (Web Text))

Max Activating Token Index: 708

Text #11

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4604. Min Act: -0.1700

Data Index: 733265 (C4 (Web Text))

Max Activating Token Index: 581

Text #12

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.5231. Min Act: -0.1700

Data Index: 378198 (C4 (Web Text))

Max Activating Token Index: 906

Text #13

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4290. Min Act: -0.1700

Data Index: 768631 (C4 (Web Text))

Max Activating Token Index: 842

Text #14

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4290. Min Act: -0.1700

Data Index: 91933 (C4 (Web Text))

Max Activating Token Index: 789

Text #15

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.3663. Min Act: -0.1700

Data Index: 1338287 (C4 (Web Text))

Max Activating Token Index: 729

Text #16

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.3977. Min Act: -0.1700

Data Index: 509402 (C4 (Web Text))

Max Activating Token Index: 973

Text #17

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4290. Min Act: -0.1700

Data Index: 1091136 (C4 (Web Text))

Max Activating Token Index: 331

Text #18

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.4290. Min Act: -0.1700

Data Index: 1020539 (C4 (Web Text))

Max Activating Token Index: 812

Text #19

Max Range: 3.8048. Min Range: -3.8048

Max Act: 3.3663. Min Act: -0.1700

Data Index: 1000045 (C4 (Web Text))

Max Activating Token Index: 996