Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1550 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.8250. Min Act: -0.1699

Data Index: 844609 (C4 (Web Text))

Max Activating Token Index: 552

Text #1

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6842. Min Act: -0.1700

Data Index: 1209143 (C4 (Web Text))

Max Activating Token Index: 48

Text #2

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.7468. Min Act: -0.1700

Data Index: 1098100 (C4 (Web Text))

Max Activating Token Index: 267

Text #3

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6842. Min Act: -0.1700

Data Index: 1313225 (C4 (Web Text))

Max Activating Token Index: 949

Text #4

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.7155. Min Act: -0.1700

Data Index: 570892 (C4 (Web Text))

Max Activating Token Index: 433

Text #5

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6686. Min Act: -0.1700

Data Index: 1129561 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6999. Min Act: -0.1700

Data Index: 146063 (C4 (Web Text))

Max Activating Token Index: 681

Text #7

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6686. Min Act: -0.1700

Data Index: 51345 (C4 (Web Text))

Max Activating Token Index: 419

Text #8

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6686. Min Act: -0.1700

Data Index: 920402 (C4 (Web Text))

Max Activating Token Index: 153

Text #9

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6842. Min Act: -0.1700

Data Index: 814433 (C4 (Web Text))

Max Activating Token Index: 799

Text #10

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6529. Min Act: -0.1700

Data Index: 338854 (C4 (Web Text))

Max Activating Token Index: 305

Text #11

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6529. Min Act: -0.1700

Data Index: 1331976 (C4 (Web Text))

Max Activating Token Index: 807

Text #12

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6373. Min Act: -0.1700

Data Index: 600786 (C4 (Web Text))

Max Activating Token Index: 938

Text #13

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6059. Min Act: -0.1700

Data Index: 989558 (C4 (Web Text))

Max Activating Token Index: 483

Text #14

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6216. Min Act: -0.1700

Data Index: 609304 (C4 (Web Text))

Max Activating Token Index: 122

Text #15

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6216. Min Act: -0.1700

Data Index: 855025 (C4 (Web Text))

Max Activating Token Index: 102

Text #16

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6216. Min Act: -0.1700

Data Index: 658552 (C4 (Web Text))

Max Activating Token Index: 1019

Text #17

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6373. Min Act: -0.1700

Data Index: 764130 (C4 (Web Text))

Max Activating Token Index: 363

Text #18

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6529. Min Act: -0.1700

Data Index: 343874 (C4 (Web Text))

Max Activating Token Index: 790

Text #19

Max Range: 3.8250. Min Range: -3.8250

Max Act: 3.6059. Min Act: -0.1700

Data Index: 396096 (C4 (Web Text))

Max Activating Token Index: 212