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

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

Neuron 8 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: 3.6822. Min Range: -3.6822

Max Act: 3.6822. Min Act: -0.1700

Data Index: 694641 (C4 (Web Text))

Max Activating Token Index: 847

Text #1

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.6196. Min Act: -0.1700

Data Index: 1045328 (C4 (Web Text))

Max Activating Token Index: 525

Text #2

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.5413. Min Act: -0.1700

Data Index: 287490 (C4 (Web Text))

Max Activating Token Index: 874

Text #3

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4943. Min Act: -0.1700

Data Index: 650272 (C4 (Web Text))

Max Activating Token Index: 181

Text #4

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4316. Min Act: -0.1700

Data Index: 792450 (C4 (Web Text))

Max Activating Token Index: 759

Text #5

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4473. Min Act: -0.1700

Data Index: 453491 (C4 (Web Text))

Max Activating Token Index: 628

Text #6

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3846. Min Act: -0.1700

Data Index: 1264093 (C4 (Web Text))

Max Activating Token Index: 317

Text #7

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4473. Min Act: -0.1700

Data Index: 681025 (C4 (Web Text))

Max Activating Token Index: 473

Text #8

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4160. Min Act: -0.1700

Data Index: 990289 (C4 (Web Text))

Max Activating Token Index: 42

Text #9

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.4003. Min Act: -0.1700

Data Index: 480784 (C4 (Web Text))

Max Activating Token Index: 857

Text #10

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3689. Min Act: -0.1700

Data Index: 588301 (C4 (Web Text))

Max Activating Token Index: 757

Text #11

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3846. Min Act: -0.1700

Data Index: 386994 (C4 (Web Text))

Max Activating Token Index: 289

Text #12

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3532. Min Act: -0.1700

Data Index: 407778 (C4 (Web Text))

Max Activating Token Index: 737

Text #13

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3846. Min Act: -0.1700

Data Index: 856366 (C4 (Web Text))

Max Activating Token Index: 327

Text #14

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3846. Min Act: -0.1700

Data Index: 129291 (C4 (Web Text))

Max Activating Token Index: 896

Text #15

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3375. Min Act: -0.1700

Data Index: 312118 (C4 (Web Text))

Max Activating Token Index: 976

Text #16

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3689. Min Act: -0.1700

Data Index: 1211385 (C4 (Web Text))

Max Activating Token Index: 730

Text #17

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3689. Min Act: -0.1700

Data Index: 941714 (C4 (Web Text))

Max Activating Token Index: 244

Text #18

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3689. Min Act: -0.1700

Data Index: 1534229 (Python Code)

Max Activating Token Index: 520

Text #19

Max Range: 3.6822. Min Range: -3.6822

Max Act: 3.3532. Min Act: -0.1700

Data Index: 84350 (C4 (Web Text))

Max Activating Token Index: 989