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

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

Neuron 700 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.8949. Min Act: -0.1700

Data Index: 1609910 (Python Code)

Max Activating Token Index: 245

Text #1

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.8949. Min Act: -0.1700

Data Index: 1634245 (Python Code)

Max Activating Token Index: 316

Text #2

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.8012. Min Act: -0.1700

Data Index: 1432592 (Python Code)

Max Activating Token Index: 554

Text #3

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.7074. Min Act: -0.1700

Data Index: 1641329 (Python Code)

Max Activating Token Index: 324

Text #4

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.6137. Min Act: -0.1700

Data Index: 1500695 (Python Code)

Max Activating Token Index: 208

Text #5

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.6137. Min Act: -0.1700

Data Index: 1603664 (Python Code)

Max Activating Token Index: 537

Text #6

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.6137. Min Act: -0.1700

Data Index: 1544499 (Python Code)

Max Activating Token Index: 478

Text #7

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.6137. Min Act: -0.1700

Data Index: 1617628 (Python Code)

Max Activating Token Index: 253

Text #8

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.6137. Min Act: -0.1700

Data Index: 1567725 (Python Code)

Max Activating Token Index: 269

Text #9

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.5824. Min Act: -0.1700

Data Index: 1422434 (Python Code)

Max Activating Token Index: 922

Text #10

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.5824. Min Act: -0.1700

Data Index: 1551016 (Python Code)

Max Activating Token Index: 149

Text #11

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.5512. Min Act: -0.1700

Data Index: 1413153 (Python Code)

Max Activating Token Index: 307

Text #12

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.5199. Min Act: -0.1700

Data Index: 1544711 (Python Code)

Max Activating Token Index: 533

Text #13

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.5512. Min Act: -0.1700

Data Index: 1433648 (Python Code)

Max Activating Token Index: 1002

Text #14

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.4574. Min Act: -0.1700

Data Index: 1455246 (Python Code)

Max Activating Token Index: 443

Text #15

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.3949. Min Act: -0.1700

Data Index: 1583864 (Python Code)

Max Activating Token Index: 73

Text #16

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.3481. Min Act: -0.1700

Data Index: 1413998 (Python Code)

Max Activating Token Index: 682

Text #17

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.4574. Min Act: -0.1700

Data Index: 35593 (C4 (Web Text))

Max Activating Token Index: 159

Text #18

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.3949. Min Act: -0.1696

Data Index: 1412961 (Python Code)

Max Activating Token Index: 473

Text #19

Max Range: 5.8949. Min Range: -5.8949

Max Act: 5.4887. Min Act: -0.1700

Data Index: 1642086 (Python Code)

Max Activating Token Index: 720