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

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

Neuron 2042 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.2615. Min Act: -0.1700

Data Index: 1484360 (Python Code)

Max Activating Token Index: 517

Text #1

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.2458. Min Act: -0.1700

Data Index: 1384150 (Python Code)

Max Activating Token Index: 974

Text #2

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.2301. Min Act: -0.1700

Data Index: 1505701 (Python Code)

Max Activating Token Index: 313

Text #3

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.1987. Min Act: -0.1700

Data Index: 1480221 (Python Code)

Max Activating Token Index: 294

Text #4

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.1514. Min Act: -0.1700

Data Index: 1303802 (C4 (Web Text))

Max Activating Token Index: 977

Text #5

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0884. Min Act: -0.1696

Data Index: 1446751 (Python Code)

Max Activating Token Index: 590

Text #6

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.1042. Min Act: -0.1700

Data Index: 1363683 (Python Code)

Max Activating Token Index: 697

Text #7

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.1199. Min Act: -0.1698

Data Index: 1486445 (Python Code)

Max Activating Token Index: 863

Text #8

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0726. Min Act: -0.1700

Data Index: 1446007 (Python Code)

Max Activating Token Index: 600

Text #9

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0726. Min Act: -0.1700

Data Index: 1530519 (Python Code)

Max Activating Token Index: 717

Text #10

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0568. Min Act: -0.1700

Data Index: 1491850 (Python Code)

Max Activating Token Index: 1004

Text #11

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0726. Min Act: -0.1700

Data Index: 1469966 (Python Code)

Max Activating Token Index: 688

Text #12

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0410. Min Act: -0.1700

Data Index: 1443671 (Python Code)

Max Activating Token Index: 712

Text #13

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0884. Min Act: -0.1700

Data Index: 1516165 (Python Code)

Max Activating Token Index: 718

Text #14

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0726. Min Act: -0.1700

Data Index: 1261145 (C4 (Web Text))

Max Activating Token Index: 962

Text #15

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0568. Min Act: -0.1700

Data Index: 1609009 (Python Code)

Max Activating Token Index: 756

Text #16

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0253. Min Act: -0.1700

Data Index: 1397258 (Python Code)

Max Activating Token Index: 411

Text #17

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0253. Min Act: -0.1699

Data Index: 1558434 (Python Code)

Max Activating Token Index: 421

Text #18

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0568. Min Act: -0.1699

Data Index: 1510888 (Python Code)

Max Activating Token Index: 902

Text #19

Max Range: 3.2615. Min Range: -3.2615

Max Act: 3.0095. Min Act: -0.1700

Data Index: 1543423 (Python Code)

Max Activating Token Index: 410