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

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

Neuron 976 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2896. Min Range: -5.2896

Max Act: 5.2896. Min Act: -0.1699

Data Index: 486290 (C4 (Web Text))

Max Activating Token Index: 968

Text #1

Max Range: 5.2896. Min Range: -5.2896

Max Act: 5.1340. Min Act: -0.1699

Data Index: 1649425 (Python Code)

Max Activating Token Index: 733

Text #2

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.9029. Min Act: -0.1699

Data Index: 64500 (C4 (Web Text))

Max Activating Token Index: 878

Text #3

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.9107. Min Act: -0.1699

Data Index: 172836 (C4 (Web Text))

Max Activating Token Index: 192

Text #4

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.8316. Min Act: -0.1699

Data Index: 1177262 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.8160. Min Act: -0.1699

Data Index: 182680 (C4 (Web Text))

Max Activating Token Index: 521

Text #6

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.8394. Min Act: -0.1699

Data Index: 1103660 (C4 (Web Text))

Max Activating Token Index: 1020

Text #7

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7593. Min Act: -0.1699

Data Index: 342400 (C4 (Web Text))

Max Activating Token Index: 921

Text #8

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7398. Min Act: -0.1699

Data Index: 787586 (C4 (Web Text))

Max Activating Token Index: 463

Text #9

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7300. Min Act: -0.1699

Data Index: 434161 (C4 (Web Text))

Max Activating Token Index: 854

Text #10

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7437. Min Act: -0.1699

Data Index: 239632 (C4 (Web Text))

Max Activating Token Index: 793

Text #11

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7671. Min Act: -0.1699

Data Index: 1097243 (C4 (Web Text))

Max Activating Token Index: 632

Text #12

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7046. Min Act: -0.1699

Data Index: 173195 (C4 (Web Text))

Max Activating Token Index: 183

Text #13

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7359. Min Act: -0.1699

Data Index: 20333 (C4 (Web Text))

Max Activating Token Index: 357

Text #14

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.6890. Min Act: -0.1699

Data Index: 90370 (C4 (Web Text))

Max Activating Token Index: 760

Text #15

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7144. Min Act: -0.1699

Data Index: 90321 (C4 (Web Text))

Max Activating Token Index: 923

Text #16

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7242. Min Act: -0.1699

Data Index: 3909 (C4 (Web Text))

Max Activating Token Index: 901

Text #17

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.7144. Min Act: -0.1699

Data Index: 490367 (C4 (Web Text))

Max Activating Token Index: 830

Text #18

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.6734. Min Act: -0.1699

Data Index: 132809 (C4 (Web Text))

Max Activating Token Index: 298

Text #19

Max Range: 5.2896. Min Range: -5.2896

Max Act: 4.6851. Min Act: -0.1699

Data Index: 564973 (C4 (Web Text))

Max Activating Token Index: 241