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

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

Neuron 1516 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.4512. Min Act: -0.1700

Data Index: 1357187 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.2950. Min Act: -0.1700

Data Index: 181613 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.2325. Min Act: -0.1699

Data Index: 689899 (C4 (Web Text))

Max Activating Token Index: 349

Text #3

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.1075. Min Act: -0.1700

Data Index: 689191 (C4 (Web Text))

Max Activating Token Index: 254

Text #4

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.1387. Min Act: -0.1700

Data Index: 756115 (C4 (Web Text))

Max Activating Token Index: 184

Text #5

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.0762. Min Act: -0.1700

Data Index: 967337 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.0450. Min Act: -0.1700

Data Index: 1356114 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9825. Min Act: -0.1700

Data Index: 1204371 (C4 (Web Text))

Max Activating Token Index: 313

Text #8

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.0137. Min Act: -0.1700

Data Index: 632894 (C4 (Web Text))

Max Activating Token Index: 683

Text #9

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.0137. Min Act: -0.1700

Data Index: 1054791 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 5.4512. Min Range: -5.4512

Max Act: 5.0137. Min Act: -0.1700

Data Index: 793449 (C4 (Web Text))

Max Activating Token Index: 914

Text #11

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9825. Min Act: -0.1698

Data Index: 161295 (C4 (Web Text))

Max Activating Token Index: 690

Text #12

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9512. Min Act: -0.1699

Data Index: 505609 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9512. Min Act: -0.1700

Data Index: 755436 (C4 (Web Text))

Max Activating Token Index: 560

Text #14

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9200. Min Act: -0.1698

Data Index: 1036541 (C4 (Web Text))

Max Activating Token Index: 725

Text #15

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9512. Min Act: -0.1700

Data Index: 268911 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9200. Min Act: -0.1700

Data Index: 182284 (C4 (Web Text))

Max Activating Token Index: 119

Text #17

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.8887. Min Act: -0.1698

Data Index: 283356 (C4 (Web Text))

Max Activating Token Index: 994

Text #18

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.9200. Min Act: -0.1700

Data Index: 49370 (C4 (Web Text))

Max Activating Token Index: 517

Text #19

Max Range: 5.4512. Min Range: -5.4512

Max Act: 4.8887. Min Act: -0.1699

Data Index: 1476375 (Python Code)

Max Activating Token Index: 13