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

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

Neuron 4 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: 5.7850. Min Range: -5.7850

Max Act: 5.7850. Min Act: -0.1700

Data Index: 1455097 (Python Code)

Max Activating Token Index: 954

Text #1

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.7538. Min Act: -0.1700

Data Index: 1407959 (Python Code)

Max Activating Token Index: 659

Text #2

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.7225. Min Act: -0.1700

Data Index: 1584523 (Python Code)

Max Activating Token Index: 720

Text #3

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1472248 (Python Code)

Max Activating Token Index: 872

Text #4

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.7850. Min Act: -0.1700

Data Index: 1576725 (Python Code)

Max Activating Token Index: 70

Text #5

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1438697 (Python Code)

Max Activating Token Index: 915

Text #6

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1640976 (Python Code)

Max Activating Token Index: 849

Text #7

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.7538. Min Act: -0.1700

Data Index: 1475382 (Python Code)

Max Activating Token Index: 175

Text #8

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.5350. Min Act: -0.1700

Data Index: 1486549 (Python Code)

Max Activating Token Index: 809

Text #9

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1576387 (Python Code)

Max Activating Token Index: 952

Text #10

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6288. Min Act: -0.1700

Data Index: 1462829 (Python Code)

Max Activating Token Index: 859

Text #11

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6288. Min Act: -0.1700

Data Index: 1465090 (Python Code)

Max Activating Token Index: 954

Text #12

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6600. Min Act: -0.1700

Data Index: 1434026 (Python Code)

Max Activating Token Index: 803

Text #13

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1362244 (Python Code)

Max Activating Token Index: 965

Text #14

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6288. Min Act: -0.1700

Data Index: 1423964 (Python Code)

Max Activating Token Index: 45

Text #15

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.5663. Min Act: -0.1700

Data Index: 1369906 (Python Code)

Max Activating Token Index: 966

Text #16

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.7225. Min Act: -0.1700

Data Index: 1519577 (Python Code)

Max Activating Token Index: 981

Text #17

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6600. Min Act: -0.1700

Data Index: 1571895 (Python Code)

Max Activating Token Index: 803

Text #18

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6288. Min Act: -0.1700

Data Index: 1401042 (Python Code)

Max Activating Token Index: 950

Text #19

Max Range: 5.7850. Min Range: -5.7850

Max Act: 5.6913. Min Act: -0.1700

Data Index: 1380150 (Python Code)

Max Activating Token Index: 807