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

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

Neuron 1648 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: 3.8289. Min Range: -3.8289

Max Act: 3.8289. Min Act: -0.1700

Data Index: 303413 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.7351. Min Act: -0.1700

Data Index: 1196518 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.6412. Min Act: -0.1700

Data Index: 344911 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.6099. Min Act: -0.1700

Data Index: 605329 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.6099. Min Act: -0.1700

Data Index: 1278908 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.5786. Min Act: -0.1700

Data Index: 720911 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.5473. Min Act: -0.1700

Data Index: 979196 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.5159. Min Act: -0.1700

Data Index: 825372 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.4846. Min Act: -0.1700

Data Index: 23752 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.4532. Min Act: -0.1700

Data Index: 26900 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.5159. Min Act: -0.1700

Data Index: 74254 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.4846. Min Act: -0.1700

Data Index: 364411 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.4532. Min Act: -0.1700

Data Index: 1212310 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3591. Min Act: -0.1700

Data Index: 1105447 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3905. Min Act: -0.1700

Data Index: 1156141 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3591. Min Act: -0.1700

Data Index: 11975 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.2963. Min Act: -0.1700

Data Index: 709509 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3277. Min Act: -0.1700

Data Index: 364104 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3277. Min Act: -0.1700

Data Index: 2842 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 3.8289. Min Range: -3.8289

Max Act: 3.3277. Min Act: -0.1700

Data Index: 472511 (C4 (Web Text))

Max Activating Token Index: 1023