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

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

Neuron 500 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: 3.1513. Min Range: -3.1513

Max Act: 3.1513. Min Act: -0.1700

Data Index: 494480 (C4 (Web Text))

Max Activating Token Index: 780

Text #1

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.1356. Min Act: -0.1700

Data Index: 840622 (C4 (Web Text))

Max Activating Token Index: 266

Text #2

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.1041. Min Act: -0.1700

Data Index: 391051 (C4 (Web Text))

Max Activating Token Index: 933

Text #3

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0883. Min Act: -0.1700

Data Index: 1332725 (C4 (Web Text))

Max Activating Token Index: 459

Text #4

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0883. Min Act: -0.1700

Data Index: 1299585 (C4 (Web Text))

Max Activating Token Index: 49

Text #5

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0725. Min Act: -0.1700

Data Index: 9130 (C4 (Web Text))

Max Activating Token Index: 298

Text #6

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0252. Min Act: -0.1700

Data Index: 884405 (C4 (Web Text))

Max Activating Token Index: 806

Text #7

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0410. Min Act: -0.1700

Data Index: 85145 (C4 (Web Text))

Max Activating Token Index: 685

Text #8

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0252. Min Act: -0.1700

Data Index: 959563 (C4 (Web Text))

Max Activating Token Index: 956

Text #9

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0252. Min Act: -0.1700

Data Index: 479927 (C4 (Web Text))

Max Activating Token Index: 79

Text #10

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 691425 (C4 (Web Text))

Max Activating Token Index: 150

Text #11

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 1083262 (C4 (Web Text))

Max Activating Token Index: 820

Text #12

Max Range: 3.1513. Min Range: -3.1513

Max Act: 3.0094. Min Act: -0.1700

Data Index: 66342 (C4 (Web Text))

Max Activating Token Index: 657

Text #13

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 692511 (C4 (Web Text))

Max Activating Token Index: 684

Text #14

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9777. Min Act: -0.1700

Data Index: 980007 (C4 (Web Text))

Max Activating Token Index: 243

Text #15

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 269519 (C4 (Web Text))

Max Activating Token Index: 1008

Text #16

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 831268 (C4 (Web Text))

Max Activating Token Index: 580

Text #17

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9936. Min Act: -0.1700

Data Index: 629500 (C4 (Web Text))

Max Activating Token Index: 133

Text #18

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9619. Min Act: -0.1700

Data Index: 790189 (C4 (Web Text))

Max Activating Token Index: 700

Text #19

Max Range: 3.1513. Min Range: -3.1513

Max Act: 2.9777. Min Act: -0.1700

Data Index: 946040 (C4 (Web Text))

Max Activating Token Index: 830