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

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

Neuron 105 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: 5.4957. Min Range: -5.4957

Max Act: 5.4957. Min Act: -0.1700

Data Index: 77873 (C4 (Web Text))

Max Activating Token Index: 325

Text #1

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4645. Min Act: -0.1700

Data Index: 358910 (C4 (Web Text))

Max Activating Token Index: 661

Text #2

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4332. Min Act: -0.1700

Data Index: 391559 (C4 (Web Text))

Max Activating Token Index: 828

Text #3

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4332. Min Act: -0.1700

Data Index: 908726 (C4 (Web Text))

Max Activating Token Index: 312

Text #4

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4332. Min Act: -0.1700

Data Index: 194668 (C4 (Web Text))

Max Activating Token Index: 978

Text #5

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 180698 (C4 (Web Text))

Max Activating Token Index: 682

Text #6

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4020. Min Act: -0.1700

Data Index: 868768 (C4 (Web Text))

Max Activating Token Index: 131

Text #7

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4020. Min Act: -0.1700

Data Index: 36135 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 4905 (C4 (Web Text))

Max Activating Token Index: 783

Text #9

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 1134446 (C4 (Web Text))

Max Activating Token Index: 388

Text #10

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4332. Min Act: -0.1700

Data Index: 1294691 (C4 (Web Text))

Max Activating Token Index: 769

Text #11

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.4020. Min Act: -0.1700

Data Index: 784434 (C4 (Web Text))

Max Activating Token Index: 916

Text #12

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 529700 (C4 (Web Text))

Max Activating Token Index: 154

Text #13

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 1124914 (C4 (Web Text))

Max Activating Token Index: 417

Text #14

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 466679 (C4 (Web Text))

Max Activating Token Index: 407

Text #15

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3395. Min Act: -0.1700

Data Index: 505624 (C4 (Web Text))

Max Activating Token Index: 307

Text #16

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3395. Min Act: -0.1700

Data Index: 558827 (C4 (Web Text))

Max Activating Token Index: 164

Text #17

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3395. Min Act: -0.1700

Data Index: 592160 (C4 (Web Text))

Max Activating Token Index: 110

Text #18

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3082. Min Act: -0.1700

Data Index: 37809 (C4 (Web Text))

Max Activating Token Index: 183

Text #19

Max Range: 5.4957. Min Range: -5.4957

Max Act: 5.3707. Min Act: -0.1700

Data Index: 277842 (C4 (Web Text))

Max Activating Token Index: 294