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

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

Neuron 618 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3705. Min Range: -5.3705

Max Act: 5.3705. Min Act: -0.1700

Data Index: 1343149 (C4 (Web Text))

Max Activating Token Index: 166

Text #1

Max Range: 5.3705. Min Range: -5.3705

Max Act: 5.0268. Min Act: -0.1700

Data Index: 259371 (C4 (Web Text))

Max Activating Token Index: 417

Text #2

Max Range: 5.3705. Min Range: -5.3705

Max Act: 5.0268. Min Act: -0.1700

Data Index: 1223415 (C4 (Web Text))

Max Activating Token Index: 922

Text #3

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.9955. Min Act: -0.1700

Data Index: 620188 (C4 (Web Text))

Max Activating Token Index: 954

Text #4

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.8393. Min Act: -0.1699

Data Index: 855269 (C4 (Web Text))

Max Activating Token Index: 85

Text #5

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7768. Min Act: -0.1700

Data Index: 774110 (C4 (Web Text))

Max Activating Token Index: 695

Text #6

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7455. Min Act: -0.1700

Data Index: 371011 (C4 (Web Text))

Max Activating Token Index: 856

Text #7

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7455. Min Act: -0.1700

Data Index: 992483 (C4 (Web Text))

Max Activating Token Index: 452

Text #8

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7143. Min Act: -0.1700

Data Index: 167260 (C4 (Web Text))

Max Activating Token Index: 41

Text #9

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7143. Min Act: -0.1699

Data Index: 62872 (C4 (Web Text))

Max Activating Token Index: 209

Text #10

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6830. Min Act: -0.1700

Data Index: 1351446 (C4 (Web Text))

Max Activating Token Index: 954

Text #11

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.7143. Min Act: -0.1700

Data Index: 505211 (C4 (Web Text))

Max Activating Token Index: 1002

Text #12

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6830. Min Act: -0.1700

Data Index: 9382 (C4 (Web Text))

Max Activating Token Index: 103

Text #13

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6518. Min Act: -0.1700

Data Index: 282186 (C4 (Web Text))

Max Activating Token Index: 317

Text #14

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6205. Min Act: -0.1700

Data Index: 1063114 (C4 (Web Text))

Max Activating Token Index: 427

Text #15

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6830. Min Act: -0.1700

Data Index: 320885 (C4 (Web Text))

Max Activating Token Index: 645

Text #16

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6830. Min Act: -0.1700

Data Index: 171042 (C4 (Web Text))

Max Activating Token Index: 574

Text #17

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6205. Min Act: -0.1699

Data Index: 668020 (C4 (Web Text))

Max Activating Token Index: 100

Text #18

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.6205. Min Act: -0.1700

Data Index: 670875 (C4 (Web Text))

Max Activating Token Index: 37

Text #19

Max Range: 5.3705. Min Range: -5.3705

Max Act: 4.5580. Min Act: -0.1700

Data Index: 1126872 (C4 (Web Text))

Max Activating Token Index: 889