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

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

Neuron 560 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.8934. Min Act: -0.1700

Data Index: 912887 (C4 (Web Text))

Max Activating Token Index: 422

Text #1

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.7684. Min Act: -0.1700

Data Index: 666403 (C4 (Web Text))

Max Activating Token Index: 833

Text #2

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.7997. Min Act: -0.1700

Data Index: 892907 (C4 (Web Text))

Max Activating Token Index: 555

Text #3

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.6747. Min Act: -0.1700

Data Index: 610238 (C4 (Web Text))

Max Activating Token Index: 941

Text #4

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5809. Min Act: -0.1700

Data Index: 609334 (C4 (Web Text))

Max Activating Token Index: 459

Text #5

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.6434. Min Act: -0.1700

Data Index: 101645 (C4 (Web Text))

Max Activating Token Index: 835

Text #6

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5497. Min Act: -0.1700

Data Index: 269387 (C4 (Web Text))

Max Activating Token Index: 532

Text #7

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5809. Min Act: -0.1700

Data Index: 451631 (C4 (Web Text))

Max Activating Token Index: 463

Text #8

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5809. Min Act: -0.1700

Data Index: 766122 (C4 (Web Text))

Max Activating Token Index: 174

Text #9

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5809. Min Act: -0.1700

Data Index: 672888 (C4 (Web Text))

Max Activating Token Index: 989

Text #10

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5497. Min Act: -0.1700

Data Index: 735118 (C4 (Web Text))

Max Activating Token Index: 375

Text #11

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5184. Min Act: -0.1700

Data Index: 1134969 (C4 (Web Text))

Max Activating Token Index: 224

Text #12

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5184. Min Act: -0.1700

Data Index: 183020 (C4 (Web Text))

Max Activating Token Index: 85

Text #13

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.5184. Min Act: -0.1700

Data Index: 626239 (C4 (Web Text))

Max Activating Token Index: 400

Text #14

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.4247. Min Act: -0.1700

Data Index: 1127941 (C4 (Web Text))

Max Activating Token Index: 887

Text #15

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.4559. Min Act: -0.1700

Data Index: 617721 (C4 (Web Text))

Max Activating Token Index: 49

Text #16

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.4559. Min Act: -0.1700

Data Index: 89743 (C4 (Web Text))

Max Activating Token Index: 211

Text #17

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.4247. Min Act: -0.1700

Data Index: 1341164 (C4 (Web Text))

Max Activating Token Index: 909

Text #18

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.3934. Min Act: -0.1700

Data Index: 122124 (C4 (Web Text))

Max Activating Token Index: 327

Text #19

Max Range: 5.8934. Min Range: -5.8934

Max Act: 5.4559. Min Act: -0.1700

Data Index: 323362 (C4 (Web Text))

Max Activating Token Index: 721