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

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

Neuron 1005 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.5335. Min Act: -0.1700

Data Index: 429571 (C4 (Web Text))

Max Activating Token Index: 8

Text #1

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.5173. Min Act: -0.1700

Data Index: 1284966 (C4 (Web Text))

Max Activating Token Index: 582

Text #2

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.5011. Min Act: -0.1700

Data Index: 598521 (C4 (Web Text))

Max Activating Token Index: 99

Text #3

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4849. Min Act: -0.1700

Data Index: 246711 (C4 (Web Text))

Max Activating Token Index: 15

Text #4

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.5011. Min Act: -0.1700

Data Index: 51424 (C4 (Web Text))

Max Activating Token Index: 81

Text #5

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4687. Min Act: -0.1700

Data Index: 1026150 (C4 (Web Text))

Max Activating Token Index: 21

Text #6

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 701889 (C4 (Web Text))

Max Activating Token Index: 280

Text #7

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4687. Min Act: -0.1700

Data Index: 792018 (C4 (Web Text))

Max Activating Token Index: 619

Text #8

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 415387 (C4 (Web Text))

Max Activating Token Index: 203

Text #9

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 1351129 (C4 (Web Text))

Max Activating Token Index: 413

Text #10

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4362. Min Act: -0.1700

Data Index: 1147007 (C4 (Web Text))

Max Activating Token Index: 181

Text #11

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 884882 (C4 (Web Text))

Max Activating Token Index: 706

Text #12

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 893904 (C4 (Web Text))

Max Activating Token Index: 6

Text #13

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 994998 (C4 (Web Text))

Max Activating Token Index: 814

Text #14

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4362. Min Act: -0.1700

Data Index: 750917 (C4 (Web Text))

Max Activating Token Index: 993

Text #15

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4362. Min Act: -0.1700

Data Index: 1249451 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4199. Min Act: -0.1700

Data Index: 826992 (C4 (Web Text))

Max Activating Token Index: 46

Text #17

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4362. Min Act: -0.1700

Data Index: 328399 (C4 (Web Text))

Max Activating Token Index: 494

Text #18

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4362. Min Act: -0.1700

Data Index: 24117 (C4 (Web Text))

Max Activating Token Index: 475

Text #19

Max Range: 2.5335. Min Range: -2.5335

Max Act: 2.4525. Min Act: -0.1700

Data Index: 506308 (C4 (Web Text))

Max Activating Token Index: 23