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

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

Neuron 961 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0489. Min Range: -3.0489

Max Act: 3.0489. Min Act: -0.1700

Data Index: 1634055 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.0489. Min Range: -3.0489

Max Act: 3.0015. Min Act: -0.1700

Data Index: 1582354 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9857. Min Act: -0.1700

Data Index: 200899 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9699. Min Act: -0.1700

Data Index: 1507372 (Python Code)

Max Activating Token Index: 314

Text #4

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9382. Min Act: -0.1700

Data Index: 1439114 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9382. Min Act: -0.1700

Data Index: 1609614 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9540. Min Act: -0.1700

Data Index: 1589300 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9065. Min Act: -0.1700

Data Index: 1638827 (Python Code)

Max Activating Token Index: 22

Text #8

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9065. Min Act: -0.1700

Data Index: 1402136 (Python Code)

Max Activating Token Index: 699

Text #9

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9065. Min Act: -0.1700

Data Index: 153426 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8906. Min Act: -0.1700

Data Index: 1414025 (Python Code)

Max Activating Token Index: 188

Text #11

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9065. Min Act: -0.1700

Data Index: 1480822 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8748. Min Act: -0.1700

Data Index: 1566884 (Python Code)

Max Activating Token Index: 508

Text #13

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.9065. Min Act: -0.1700

Data Index: 473098 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8589. Min Act: -0.1700

Data Index: 1398809 (Python Code)

Max Activating Token Index: 55

Text #15

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8748. Min Act: -0.1700

Data Index: 1421467 (Python Code)

Max Activating Token Index: 770

Text #16

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8589. Min Act: -0.1700

Data Index: 1371827 (Python Code)

Max Activating Token Index: 23

Text #17

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8430. Min Act: -0.1700

Data Index: 1642155 (Python Code)

Max Activating Token Index: 712

Text #18

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8430. Min Act: -0.1700

Data Index: 1485699 (Python Code)

Max Activating Token Index: 10

Text #19

Max Range: 3.0489. Min Range: -3.0489

Max Act: 2.8430. Min Act: -0.1700

Data Index: 1371150 (Python Code)

Max Activating Token Index: 64