Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 800 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.2210. Min Act: -0.1700

Data Index: 1178721 (C4 (Web Text))

Max Activating Token Index: 809

Text #1

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1266. Min Act: -0.1700

Data Index: 63344 (C4 (Web Text))

Max Activating Token Index: 667

Text #2

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1581. Min Act: -0.1700

Data Index: 1178079 (C4 (Web Text))

Max Activating Token Index: 796

Text #3

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1895. Min Act: -0.1700

Data Index: 1170746 (C4 (Web Text))

Max Activating Token Index: 185

Text #4

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1423. Min Act: -0.1700

Data Index: 496215 (C4 (Web Text))

Max Activating Token Index: 300

Text #5

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1266. Min Act: -0.1700

Data Index: 594539 (C4 (Web Text))

Max Activating Token Index: 398

Text #6

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1423. Min Act: -0.1700

Data Index: 275468 (C4 (Web Text))

Max Activating Token Index: 655

Text #7

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0635. Min Act: -0.1700

Data Index: 1308230 (C4 (Web Text))

Max Activating Token Index: 845

Text #8

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0635. Min Act: -0.1700

Data Index: 824066 (C4 (Web Text))

Max Activating Token Index: 817

Text #9

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.1108. Min Act: -0.1700

Data Index: 177811 (C4 (Web Text))

Max Activating Token Index: 793

Text #10

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0793. Min Act: -0.1700

Data Index: 1078671 (C4 (Web Text))

Max Activating Token Index: 972

Text #11

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0793. Min Act: -0.1700

Data Index: 33637 (C4 (Web Text))

Max Activating Token Index: 636

Text #12

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0635. Min Act: -0.1700

Data Index: 559042 (C4 (Web Text))

Max Activating Token Index: 236

Text #13

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0793. Min Act: -0.1700

Data Index: 631279 (C4 (Web Text))

Max Activating Token Index: 421

Text #14

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0950. Min Act: -0.1700

Data Index: 1086082 (C4 (Web Text))

Max Activating Token Index: 771

Text #15

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0319. Min Act: -0.1700

Data Index: 519186 (C4 (Web Text))

Max Activating Token Index: 1005

Text #16

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0161. Min Act: -0.1700

Data Index: 189934 (C4 (Web Text))

Max Activating Token Index: 796

Text #17

Max Range: 3.2210. Min Range: -3.2210

Max Act: 2.9687. Min Act: -0.1700

Data Index: 827949 (C4 (Web Text))

Max Activating Token Index: 771

Text #18

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0003. Min Act: -0.1700

Data Index: 926412 (C4 (Web Text))

Max Activating Token Index: 548

Text #19

Max Range: 3.2210. Min Range: -3.2210

Max Act: 3.0161. Min Act: -0.1700

Data Index: 903447 (C4 (Web Text))

Max Activating Token Index: 554