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

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

Neuron 718 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: 5.6192. Min Range: -5.6192

Max Act: 5.6192. Min Act: -0.1700

Data Index: 1492931 (Python Code)

Max Activating Token Index: 687

Text #1

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.6192. Min Act: -0.1700

Data Index: 1409170 (Python Code)

Max Activating Token Index: 383

Text #2

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4942. Min Act: -0.1700

Data Index: 145860 (C4 (Web Text))

Max Activating Token Index: 313

Text #3

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.5254. Min Act: -0.1700

Data Index: 1598061 (Python Code)

Max Activating Token Index: 353

Text #4

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4317. Min Act: -0.1700

Data Index: 1445616 (Python Code)

Max Activating Token Index: 900

Text #5

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4629. Min Act: -0.1700

Data Index: 1460766 (Python Code)

Max Activating Token Index: 249

Text #6

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4004. Min Act: -0.1700

Data Index: 1613121 (Python Code)

Max Activating Token Index: 825

Text #7

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4004. Min Act: -0.1700

Data Index: 1423748 (Python Code)

Max Activating Token Index: 897

Text #8

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4629. Min Act: -0.1700

Data Index: 1372643 (Python Code)

Max Activating Token Index: 431

Text #9

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4629. Min Act: -0.1700

Data Index: 1526030 (Python Code)

Max Activating Token Index: 249

Text #10

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4942. Min Act: -0.1700

Data Index: 1633783 (Python Code)

Max Activating Token Index: 179

Text #11

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4004. Min Act: -0.1700

Data Index: 1585792 (Python Code)

Max Activating Token Index: 385

Text #12

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4317. Min Act: -0.1700

Data Index: 197354 (C4 (Web Text))

Max Activating Token Index: 278

Text #13

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4317. Min Act: -0.1700

Data Index: 1514435 (Python Code)

Max Activating Token Index: 405

Text #14

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4629. Min Act: -0.1700

Data Index: 1499763 (Python Code)

Max Activating Token Index: 188

Text #15

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.2754. Min Act: -0.1700

Data Index: 1622832 (Python Code)

Max Activating Token Index: 919

Text #16

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.3066. Min Act: -0.1700

Data Index: 1507128 (Python Code)

Max Activating Token Index: 720

Text #17

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.4004. Min Act: -0.1700

Data Index: 1512844 (Python Code)

Max Activating Token Index: 253

Text #18

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.3066. Min Act: -0.1700

Data Index: 1454022 (Python Code)

Max Activating Token Index: 68

Text #19

Max Range: 5.6192. Min Range: -5.6192

Max Act: 5.2754. Min Act: -0.1700

Data Index: 1535417 (Python Code)

Max Activating Token Index: 268