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

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

Neuron 819 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: 3.2096. Min Range: -3.2096

Max Act: 3.2096. Min Act: -0.1700

Data Index: 1618689 (Python Code)

Max Activating Token Index: 3

Text #1

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.1939. Min Act: -0.1700

Data Index: 1455093 (Python Code)

Max Activating Token Index: 6

Text #2

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.1309. Min Act: -0.1699

Data Index: 1463872 (Python Code)

Max Activating Token Index: 11

Text #3

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.1309. Min Act: -0.1700

Data Index: 1621794 (Python Code)

Max Activating Token Index: 47

Text #4

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.1467. Min Act: -0.1700

Data Index: 1476469 (Python Code)

Max Activating Token Index: 38

Text #5

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.1152. Min Act: -0.1700

Data Index: 1449672 (Python Code)

Max Activating Token Index: 58

Text #6

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0994. Min Act: -0.1700

Data Index: 1446243 (Python Code)

Max Activating Token Index: 87

Text #7

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0836. Min Act: -0.1700

Data Index: 1627900 (Python Code)

Max Activating Token Index: 4

Text #8

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0679. Min Act: -0.1700

Data Index: 1457049 (Python Code)

Max Activating Token Index: 268

Text #9

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0994. Min Act: -0.1700

Data Index: 1592173 (Python Code)

Max Activating Token Index: 224

Text #10

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0679. Min Act: -0.1700

Data Index: 1612434 (Python Code)

Max Activating Token Index: 89

Text #11

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0679. Min Act: -0.1700

Data Index: 1465685 (Python Code)

Max Activating Token Index: 459

Text #12

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0679. Min Act: -0.1700

Data Index: 1647396 (Python Code)

Max Activating Token Index: 893

Text #13

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0679. Min Act: -0.1700

Data Index: 1597037 (Python Code)

Max Activating Token Index: 19

Text #14

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0521. Min Act: -0.1700

Data Index: 1624112 (Python Code)

Max Activating Token Index: 556

Text #15

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0521. Min Act: -0.1700

Data Index: 1652961 (Python Code)

Max Activating Token Index: 38

Text #16

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0521. Min Act: -0.1700

Data Index: 1496036 (Python Code)

Max Activating Token Index: 709

Text #17

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0521. Min Act: -0.1700

Data Index: 1575840 (Python Code)

Max Activating Token Index: 726

Text #18

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0521. Min Act: -0.1700

Data Index: 1533057 (Python Code)

Max Activating Token Index: 399

Text #19

Max Range: 3.2096. Min Range: -3.2096

Max Act: 3.0363. Min Act: -0.1700

Data Index: 1394876 (Python Code)

Max Activating Token Index: 186