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

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

Neuron 168 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: 3.9126. Min Range: -3.9126

Max Act: 3.9126. Min Act: -0.1700

Data Index: 288855 (C4 (Web Text))

Max Activating Token Index: 324

Text #1

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.6780. Min Act: -0.1700

Data Index: 1490303 (Python Code)

Max Activating Token Index: 266

Text #2

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.6310. Min Act: -0.1700

Data Index: 652519 (C4 (Web Text))

Max Activating Token Index: 57

Text #3

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5841. Min Act: -0.1700

Data Index: 1019165 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5841. Min Act: -0.1700

Data Index: 1613077 (Python Code)

Max Activating Token Index: 408

Text #5

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5684. Min Act: -0.1700

Data Index: 264212 (C4 (Web Text))

Max Activating Token Index: 418

Text #6

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5527. Min Act: -0.1700

Data Index: 1652604 (Python Code)

Max Activating Token Index: 360

Text #7

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5214. Min Act: -0.1700

Data Index: 1645293 (Python Code)

Max Activating Token Index: 432

Text #8

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4431. Min Act: -0.1700

Data Index: 1186910 (C4 (Web Text))

Max Activating Token Index: 342

Text #9

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4587. Min Act: -0.1700

Data Index: 1632798 (Python Code)

Max Activating Token Index: 703

Text #10

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 1312328 (C4 (Web Text))

Max Activating Token Index: 868

Text #11

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 1109352 (C4 (Web Text))

Max Activating Token Index: 996

Text #12

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.5214. Min Act: -0.1700

Data Index: 41944 (C4 (Web Text))

Max Activating Token Index: 975

Text #13

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 1314443 (C4 (Web Text))

Max Activating Token Index: 901

Text #14

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 212177 (C4 (Web Text))

Max Activating Token Index: 780

Text #15

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 1580264 (Python Code)

Max Activating Token Index: 876

Text #16

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4901. Min Act: -0.1700

Data Index: 1565222 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 789519 (C4 (Web Text))

Max Activating Token Index: 816

Text #18

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4744. Min Act: -0.1700

Data Index: 152614 (C4 (Web Text))

Max Activating Token Index: 645

Text #19

Max Range: 3.9126. Min Range: -3.9126

Max Act: 3.4274. Min Act: -0.1700

Data Index: 1034315 (C4 (Web Text))

Max Activating Token Index: 620