Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1735 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2860. Min Act: -0.1135

Data Index: 1070520 (C4 (Web Text))

Max Activating Token Index: 246

Text #1

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2744. Min Act: -0.1656

Data Index: 1491673 (Python Code)

Max Activating Token Index: 662

Text #2

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2923. Min Act: -0.1339

Data Index: 1575301 (Python Code)

Max Activating Token Index: 55

Text #3

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2774. Min Act: -0.1694

Data Index: 1434319 (Python Code)

Max Activating Token Index: 263

Text #4

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2851. Min Act: -0.1699

Data Index: 70596 (C4 (Web Text))

Max Activating Token Index: 916

Text #5

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2708. Min Act: -0.1700

Data Index: 898878 (C4 (Web Text))

Max Activating Token Index: 472

Text #6

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2676. Min Act: -0.1699

Data Index: 1504559 (Python Code)

Max Activating Token Index: 227

Text #7

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2389. Min Act: -0.1700

Data Index: 865337 (C4 (Web Text))

Max Activating Token Index: 928

Text #8

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2150. Min Act: -0.1699

Data Index: 1632011 (Python Code)

Max Activating Token Index: 548

Text #9

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2182. Min Act: -0.1135

Data Index: 1547083 (Python Code)

Max Activating Token Index: 929

Text #10

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1878. Min Act: -0.1700

Data Index: 1304183 (C4 (Web Text))

Max Activating Token Index: 980

Text #11

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2030. Min Act: -0.1673

Data Index: 1563817 (Python Code)

Max Activating Token Index: 1010

Text #12

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1868. Min Act: -0.1691

Data Index: 1435994 (Python Code)

Max Activating Token Index: 907

Text #13

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1892. Min Act: -0.1684

Data Index: 1363431 (Python Code)

Max Activating Token Index: 621

Text #14

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1922. Min Act: -0.1454

Data Index: 1618821 (Python Code)

Max Activating Token Index: 597

Text #15

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.2091. Min Act: -0.1659

Data Index: 1426660 (Python Code)

Max Activating Token Index: 188

Text #16

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1786. Min Act: -0.1699

Data Index: 364267 (C4 (Web Text))

Max Activating Token Index: 966

Text #17

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1607. Min Act: -0.1699

Data Index: 1435664 (Python Code)

Max Activating Token Index: 884

Text #18

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1521. Min Act: -0.1626

Data Index: 1600998 (Python Code)

Max Activating Token Index: 527

Text #19

Max Range: 3.2923. Min Range: -3.2923

Max Act: 3.1208. Min Act: -0.1696

Data Index: 1586005 (Python Code)

Max Activating Token Index: 970