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

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

Neuron 707 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: 7.4271. Min Range: -7.4271

Max Act: 7.4271. Min Act: -0.1700

Data Index: 691557 (C4 (Web Text))

Max Activating Token Index: 976

Text #1

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.8959. Min Act: -0.1700

Data Index: 74405 (C4 (Web Text))

Max Activating Token Index: 702

Text #2

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.7084. Min Act: -0.1700

Data Index: 1147716 (C4 (Web Text))

Max Activating Token Index: 791

Text #3

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.5209. Min Act: -0.1700

Data Index: 979543 (C4 (Web Text))

Max Activating Token Index: 986

Text #4

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.4584. Min Act: -0.1700

Data Index: 1131228 (C4 (Web Text))

Max Activating Token Index: 112

Text #5

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3959. Min Act: -0.1700

Data Index: 695877 (C4 (Web Text))

Max Activating Token Index: 434

Text #6

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3646. Min Act: -0.1700

Data Index: 932571 (C4 (Web Text))

Max Activating Token Index: 315

Text #7

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3646. Min Act: -0.1700

Data Index: 589982 (C4 (Web Text))

Max Activating Token Index: 300

Text #8

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3646. Min Act: -0.1700

Data Index: 1071449 (C4 (Web Text))

Max Activating Token Index: 536

Text #9

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3334. Min Act: -0.1700

Data Index: 1002602 (C4 (Web Text))

Max Activating Token Index: 763

Text #10

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.2709. Min Act: -0.1700

Data Index: 1335182 (C4 (Web Text))

Max Activating Token Index: 458

Text #11

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.2709. Min Act: -0.1700

Data Index: 953688 (C4 (Web Text))

Max Activating Token Index: 963

Text #12

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.3021. Min Act: -0.1700

Data Index: 45383 (C4 (Web Text))

Max Activating Token Index: 660

Text #13

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.2709. Min Act: -0.1700

Data Index: 1267082 (C4 (Web Text))

Max Activating Token Index: 485

Text #14

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.2084. Min Act: -0.1700

Data Index: 280070 (C4 (Web Text))

Max Activating Token Index: 737

Text #15

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.1459. Min Act: -0.1700

Data Index: 632661 (C4 (Web Text))

Max Activating Token Index: 511

Text #16

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.1771. Min Act: -0.1700

Data Index: 1309503 (C4 (Web Text))

Max Activating Token Index: 683

Text #17

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.1459. Min Act: -0.1700

Data Index: 945472 (C4 (Web Text))

Max Activating Token Index: 110

Text #18

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.1771. Min Act: -0.1700

Data Index: 785022 (C4 (Web Text))

Max Activating Token Index: 120

Text #19

Max Range: 7.4271. Min Range: -7.4271

Max Act: 6.1459. Min Act: -0.1700

Data Index: 876156 (C4 (Web Text))

Max Activating Token Index: 782