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

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

Neuron 594 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.6295. Min Act: -0.1700

Data Index: 1393753 (Python Code)

Max Activating Token Index: 719

Text #1

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.5042. Min Act: -0.1700

Data Index: 1443849 (Python Code)

Max Activating Token Index: 999

Text #2

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.3003. Min Act: -0.1700

Data Index: 1029820 (C4 (Web Text))

Max Activating Token Index: 854

Text #3

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2846. Min Act: -0.1700

Data Index: 488020 (C4 (Web Text))

Max Activating Token Index: 998

Text #4

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2688. Min Act: -0.1700

Data Index: 1604470 (Python Code)

Max Activating Token Index: 191

Text #5

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2217. Min Act: -0.1700

Data Index: 1614958 (Python Code)

Max Activating Token Index: 534

Text #6

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2531. Min Act: -0.1700

Data Index: 423872 (C4 (Web Text))

Max Activating Token Index: 772

Text #7

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2374. Min Act: -0.1700

Data Index: 570513 (C4 (Web Text))

Max Activating Token Index: 920

Text #8

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.2060. Min Act: -0.1700

Data Index: 1387927 (Python Code)

Max Activating Token Index: 459

Text #9

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1902. Min Act: -0.1700

Data Index: 1341067 (C4 (Web Text))

Max Activating Token Index: 787

Text #10

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1587. Min Act: -0.1700

Data Index: 1650175 (Python Code)

Max Activating Token Index: 488

Text #11

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1587. Min Act: -0.1700

Data Index: 1523291 (Python Code)

Max Activating Token Index: 876

Text #12

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1430. Min Act: -0.1700

Data Index: 1628104 (Python Code)

Max Activating Token Index: 62

Text #13

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1115. Min Act: -0.1700

Data Index: 1395453 (Python Code)

Max Activating Token Index: 27

Text #14

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1272. Min Act: -0.1700

Data Index: 1415566 (Python Code)

Max Activating Token Index: 102

Text #15

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1272. Min Act: -0.1700

Data Index: 1577867 (Python Code)

Max Activating Token Index: 563

Text #16

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1272. Min Act: -0.1700

Data Index: 820485 (C4 (Web Text))

Max Activating Token Index: 242

Text #17

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.1115. Min Act: -0.1700

Data Index: 622529 (C4 (Web Text))

Max Activating Token Index: 897

Text #18

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.0957. Min Act: -0.1700

Data Index: 1453603 (Python Code)

Max Activating Token Index: 721

Text #19

Max Range: 3.6295. Min Range: -3.6295

Max Act: 3.0957. Min Act: -0.1700

Data Index: 1369163 (Python Code)

Max Activating Token Index: 510