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

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

Neuron 496 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: 3.4334. Min Range: -3.4334

Max Act: 3.4334. Min Act: -0.1700

Data Index: 1308803 (C4 (Web Text))

Max Activating Token Index: 48

Text #1

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.2293. Min Act: -0.1700

Data Index: 1205428 (C4 (Web Text))

Max Activating Token Index: 942

Text #2

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.2293. Min Act: -0.1700

Data Index: 455410 (C4 (Web Text))

Max Activating Token Index: 57

Text #3

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1822. Min Act: -0.1700

Data Index: 7255 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1664. Min Act: -0.1700

Data Index: 336321 (C4 (Web Text))

Max Activating Token Index: 15

Text #5

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1192. Min Act: -0.1700

Data Index: 1356298 (C4 (Web Text))

Max Activating Token Index: 831

Text #6

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1192. Min Act: -0.1700

Data Index: 1132328 (C4 (Web Text))

Max Activating Token Index: 188

Text #7

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1192. Min Act: -0.1700

Data Index: 1090310 (C4 (Web Text))

Max Activating Token Index: 253

Text #8

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0876. Min Act: -0.1700

Data Index: 417031 (C4 (Web Text))

Max Activating Token Index: 131

Text #9

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.1034. Min Act: -0.1699

Data Index: 994239 (C4 (Web Text))

Max Activating Token Index: 21

Text #10

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0718. Min Act: -0.1700

Data Index: 1183472 (C4 (Web Text))

Max Activating Token Index: 77

Text #11

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0718. Min Act: -0.1700

Data Index: 628844 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0403. Min Act: -0.1700

Data Index: 1185387 (C4 (Web Text))

Max Activating Token Index: 508

Text #13

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0561. Min Act: -0.1700

Data Index: 628039 (C4 (Web Text))

Max Activating Token Index: 831

Text #14

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0245. Min Act: -0.1700

Data Index: 562485 (C4 (Web Text))

Max Activating Token Index: 172

Text #15

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0245. Min Act: -0.1700

Data Index: 1139214 (C4 (Web Text))

Max Activating Token Index: 620

Text #16

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0087. Min Act: -0.1698

Data Index: 805083 (C4 (Web Text))

Max Activating Token Index: 40

Text #17

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0245. Min Act: -0.1700

Data Index: 563779 (C4 (Web Text))

Max Activating Token Index: 243

Text #18

Max Range: 3.4334. Min Range: -3.4334

Max Act: 3.0245. Min Act: -0.1700

Data Index: 464397 (C4 (Web Text))

Max Activating Token Index: 160

Text #19

Max Range: 3.4334. Min Range: -3.4334

Max Act: 2.9929. Min Act: -0.1700

Data Index: 916761 (C4 (Web Text))

Max Activating Token Index: 55