Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 548 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.1976. Min Act: -0.1700

Data Index: 1393746 (Python Code)

Max Activating Token Index: 427

Text #1

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.1477. Min Act: -0.1700

Data Index: 1564450 (Python Code)

Max Activating Token Index: 556

Text #2

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.1144. Min Act: -0.1699

Data Index: 1418818 (Python Code)

Max Activating Token Index: 574

Text #3

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.1144. Min Act: -0.1699

Data Index: 1422825 (Python Code)

Max Activating Token Index: 933

Text #4

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.0893. Min Act: -0.1700

Data Index: 878832 (C4 (Web Text))

Max Activating Token Index: 693

Text #5

Max Range: 2.1976. Min Range: -2.1976

Max Act: 2.0641. Min Act: -0.1700

Data Index: 761643 (C4 (Web Text))

Max Activating Token Index: 364

Text #6

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9629. Min Act: -0.1700

Data Index: 1594773 (Python Code)

Max Activating Token Index: 468

Text #7

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9629. Min Act: -0.1700

Data Index: 218440 (C4 (Web Text))

Max Activating Token Index: 822

Text #8

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9629. Min Act: -0.1700

Data Index: 1536933 (Python Code)

Max Activating Token Index: 177

Text #9

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9375. Min Act: -0.1700

Data Index: 781205 (C4 (Web Text))

Max Activating Token Index: 680

Text #10

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9120. Min Act: -0.1700

Data Index: 1421097 (Python Code)

Max Activating Token Index: 503

Text #11

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9290. Min Act: -0.1700

Data Index: 995987 (C4 (Web Text))

Max Activating Token Index: 745

Text #12

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9035. Min Act: -0.1700

Data Index: 1607030 (Python Code)

Max Activating Token Index: 168

Text #13

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.9035. Min Act: -0.1700

Data Index: 1483006 (Python Code)

Max Activating Token Index: 436

Text #14

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8864. Min Act: -0.1700

Data Index: 1482165 (Python Code)

Max Activating Token Index: 387

Text #15

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8864. Min Act: -0.1700

Data Index: 1517567 (Python Code)

Max Activating Token Index: 805

Text #16

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8779. Min Act: -0.1700

Data Index: 672717 (C4 (Web Text))

Max Activating Token Index: 673

Text #17

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8694. Min Act: -0.1700

Data Index: 1536382 (Python Code)

Max Activating Token Index: 352

Text #18

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8608. Min Act: -0.1700

Data Index: 1625577 (Python Code)

Max Activating Token Index: 102

Text #19

Max Range: 2.1976. Min Range: -2.1976

Max Act: 1.8608. Min Act: -0.1700

Data Index: 1366041 (Python Code)

Max Activating Token Index: 677