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

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

Neuron 1703 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.8155. Min Act: -0.1700

Data Index: 1113387 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.7060. Min Act: -0.1700

Data Index: 1349955 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.5495. Min Act: -0.1700

Data Index: 29245 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.5651. Min Act: -0.1700

Data Index: 1394670 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.4711. Min Act: -0.1700

Data Index: 1323501 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.4398. Min Act: -0.1700

Data Index: 600484 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.4398. Min Act: -0.1700

Data Index: 80203 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3770. Min Act: -0.1700

Data Index: 552634 (C4 (Web Text))

Max Activating Token Index: 454

Text #8

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3613. Min Act: -0.1700

Data Index: 1168084 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3613. Min Act: -0.1700

Data Index: 1548046 (Python Code)

Max Activating Token Index: 18

Text #10

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2985. Min Act: -0.1700

Data Index: 1373730 (Python Code)

Max Activating Token Index: 105

Text #11

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2671. Min Act: -0.1700

Data Index: 1640276 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3770. Min Act: -0.1700

Data Index: 1392939 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3300. Min Act: -0.1700

Data Index: 279826 (C4 (Web Text))

Max Activating Token Index: 541

Text #14

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3300. Min Act: -0.1700

Data Index: 1016135 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2985. Min Act: -0.1700

Data Index: 1617852 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2828. Min Act: -0.1700

Data Index: 1438151 (Python Code)

Max Activating Token Index: 786

Text #17

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.3143. Min Act: -0.1700

Data Index: 1609030 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2828. Min Act: -0.1700

Data Index: 884446 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 3.8155. Min Range: -3.8155

Max Act: 3.2514. Min Act: -0.1700

Data Index: 787853 (C4 (Web Text))

Max Activating Token Index: 1023