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

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

Neuron 1725 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.0467. Min Range: -4.0467

Max Act: 4.0467. Min Act: -0.1700

Data Index: 1341087 (C4 (Web Text))

Max Activating Token Index: 84

Text #1

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.9216. Min Act: -0.1700

Data Index: 543729 (C4 (Web Text))

Max Activating Token Index: 864

Text #2

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.8278. Min Act: -0.1700

Data Index: 1619561 (Python Code)

Max Activating Token Index: 867

Text #3

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1498243 (Python Code)

Max Activating Token Index: 904

Text #4

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.8278. Min Act: -0.1700

Data Index: 256408 (C4 (Web Text))

Max Activating Token Index: 193

Text #5

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.8278. Min Act: -0.1700

Data Index: 1187745 (C4 (Web Text))

Max Activating Token Index: 815

Text #6

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1412167 (Python Code)

Max Activating Token Index: 997

Text #7

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7652. Min Act: -0.1700

Data Index: 1571886 (Python Code)

Max Activating Token Index: 402

Text #8

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1653219 (Python Code)

Max Activating Token Index: 867

Text #9

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1389731 (Python Code)

Max Activating Token Index: 87

Text #10

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7339. Min Act: -0.1700

Data Index: 1571600 (Python Code)

Max Activating Token Index: 985

Text #11

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1570869 (Python Code)

Max Activating Token Index: 650

Text #12

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1542977 (Python Code)

Max Activating Token Index: 1007

Text #13

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7652. Min Act: -0.1700

Data Index: 1537662 (Python Code)

Max Activating Token Index: 986

Text #14

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7339. Min Act: -0.1700

Data Index: 1367376 (Python Code)

Max Activating Token Index: 843

Text #15

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7965. Min Act: -0.1700

Data Index: 1459763 (Python Code)

Max Activating Token Index: 788

Text #16

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7652. Min Act: -0.1700

Data Index: 1425169 (Python Code)

Max Activating Token Index: 788

Text #17

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7652. Min Act: -0.1700

Data Index: 1434792 (Python Code)

Max Activating Token Index: 323

Text #18

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7652. Min Act: -0.1700

Data Index: 1378987 (Python Code)

Max Activating Token Index: 650

Text #19

Max Range: 4.0467. Min Range: -4.0467

Max Act: 3.7339. Min Act: -0.1700

Data Index: 1447462 (Python Code)

Max Activating Token Index: 782