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

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

Neuron 1168 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.4322. Min Act: -0.1700

Data Index: 1345578 (C4 (Web Text))

Max Activating Token Index: 820

Text #1

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.2915. Min Act: -0.1700

Data Index: 761811 (C4 (Web Text))

Max Activating Token Index: 834

Text #2

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.3228. Min Act: -0.1700

Data Index: 1371048 (Python Code)

Max Activating Token Index: 112

Text #3

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.3072. Min Act: -0.1700

Data Index: 754775 (C4 (Web Text))

Max Activating Token Index: 665

Text #4

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1978. Min Act: -0.1700

Data Index: 626714 (C4 (Web Text))

Max Activating Token Index: 1003

Text #5

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1665. Min Act: -0.1700

Data Index: 681690 (C4 (Web Text))

Max Activating Token Index: 753

Text #6

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1196. Min Act: -0.1699

Data Index: 1564848 (Python Code)

Max Activating Token Index: 328

Text #7

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1352. Min Act: -0.1700

Data Index: 1559191 (Python Code)

Max Activating Token Index: 347

Text #8

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1196. Min Act: -0.1700

Data Index: 981148 (C4 (Web Text))

Max Activating Token Index: 531

Text #9

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1196. Min Act: -0.1700

Data Index: 1221011 (C4 (Web Text))

Max Activating Token Index: 552

Text #10

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1196. Min Act: -0.1700

Data Index: 1643739 (Python Code)

Max Activating Token Index: 717

Text #11

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1040. Min Act: -0.1700

Data Index: 1185714 (C4 (Web Text))

Max Activating Token Index: 996

Text #12

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1196. Min Act: -0.1700

Data Index: 250404 (C4 (Web Text))

Max Activating Token Index: 610

Text #13

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0883. Min Act: -0.1700

Data Index: 1430027 (Python Code)

Max Activating Token Index: 74

Text #14

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.1040. Min Act: -0.1700

Data Index: 1400469 (Python Code)

Max Activating Token Index: 609

Text #15

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0883. Min Act: -0.1700

Data Index: 1559427 (Python Code)

Max Activating Token Index: 247

Text #16

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0727. Min Act: -0.1700

Data Index: 1359846 (Python Code)

Max Activating Token Index: 972

Text #17

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0883. Min Act: -0.1700

Data Index: 1348162 (C4 (Web Text))

Max Activating Token Index: 340

Text #18

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0571. Min Act: -0.1700

Data Index: 180538 (C4 (Web Text))

Max Activating Token Index: 115

Text #19

Max Range: 4.4322. Min Range: -4.4322

Max Act: 4.0571. Min Act: -0.1700

Data Index: 326747 (C4 (Web Text))

Max Activating Token Index: 743