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

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

Neuron 888 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: 3.1429. Min Range: -3.1429

Max Act: 3.1429. Min Act: -0.1700

Data Index: 788871 (C4 (Web Text))

Max Activating Token Index: 690

Text #1

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0956. Min Act: -0.1700

Data Index: 446414 (C4 (Web Text))

Max Activating Token Index: 123

Text #2

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0641. Min Act: -0.1700

Data Index: 300673 (C4 (Web Text))

Max Activating Token Index: 291

Text #3

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0641. Min Act: -0.1700

Data Index: 183588 (C4 (Web Text))

Max Activating Token Index: 170

Text #4

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0483. Min Act: -0.1700

Data Index: 702178 (C4 (Web Text))

Max Activating Token Index: 371

Text #5

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0167. Min Act: -0.1700

Data Index: 122475 (C4 (Web Text))

Max Activating Token Index: 184

Text #6

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0167. Min Act: -0.1700

Data Index: 994073 (C4 (Web Text))

Max Activating Token Index: 160

Text #7

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0325. Min Act: -0.1700

Data Index: 1047434 (C4 (Web Text))

Max Activating Token Index: 152

Text #8

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0325. Min Act: -0.1700

Data Index: 303914 (C4 (Web Text))

Max Activating Token Index: 145

Text #9

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9851. Min Act: -0.1700

Data Index: 1134625 (C4 (Web Text))

Max Activating Token Index: 998

Text #10

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9851. Min Act: -0.1700

Data Index: 803449 (C4 (Web Text))

Max Activating Token Index: 479

Text #11

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9851. Min Act: -0.1700

Data Index: 621855 (C4 (Web Text))

Max Activating Token Index: 128

Text #12

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0325. Min Act: -0.1700

Data Index: 1017038 (C4 (Web Text))

Max Activating Token Index: 827

Text #13

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9851. Min Act: -0.1700

Data Index: 1608469 (Python Code)

Max Activating Token Index: 300

Text #14

Max Range: 3.1429. Min Range: -3.1429

Max Act: 3.0009. Min Act: -0.1700

Data Index: 1092932 (C4 (Web Text))

Max Activating Token Index: 409

Text #15

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9534. Min Act: -0.1700

Data Index: 435162 (C4 (Web Text))

Max Activating Token Index: 10

Text #16

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9376. Min Act: -0.1700

Data Index: 1244928 (C4 (Web Text))

Max Activating Token Index: 259

Text #17

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9692. Min Act: -0.1700

Data Index: 298698 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9376. Min Act: -0.1700

Data Index: 456426 (C4 (Web Text))

Max Activating Token Index: 343

Text #19

Max Range: 3.1429. Min Range: -3.1429

Max Act: 2.9534. Min Act: -0.1700

Data Index: 222833 (C4 (Web Text))

Max Activating Token Index: 452