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

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

Neuron 271 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.7193. Min Act: -0.1700

Data Index: 298933 (C4 (Web Text))

Max Activating Token Index: 703

Text #1

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.3746. Min Act: -0.1700

Data Index: 882342 (C4 (Web Text))

Max Activating Token Index: 463

Text #2

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.4060. Min Act: -0.1700

Data Index: 1137093 (C4 (Web Text))

Max Activating Token Index: 1005

Text #3

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.3433. Min Act: -0.1700

Data Index: 1103084 (C4 (Web Text))

Max Activating Token Index: 464

Text #4

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.2490. Min Act: -0.1700

Data Index: 1394624 (Python Code)

Max Activating Token Index: 299

Text #5

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.2490. Min Act: -0.1700

Data Index: 1429841 (Python Code)

Max Activating Token Index: 611

Text #6

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.2176. Min Act: -0.1700

Data Index: 1399577 (Python Code)

Max Activating Token Index: 500

Text #7

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.2176. Min Act: -0.1700

Data Index: 5823 (C4 (Web Text))

Max Activating Token Index: 97

Text #8

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.1861. Min Act: -0.1700

Data Index: 691591 (C4 (Web Text))

Max Activating Token Index: 104

Text #9

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.2018. Min Act: -0.1700

Data Index: 1108533 (C4 (Web Text))

Max Activating Token Index: 49

Text #10

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.1389. Min Act: -0.1700

Data Index: 593297 (C4 (Web Text))

Max Activating Token Index: 32

Text #11

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.1389. Min Act: -0.1700

Data Index: 1220038 (C4 (Web Text))

Max Activating Token Index: 749

Text #12

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.1231. Min Act: -0.1700

Data Index: 222279 (C4 (Web Text))

Max Activating Token Index: 399

Text #13

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.1073. Min Act: -0.1700

Data Index: 1563810 (Python Code)

Max Activating Token Index: 296

Text #14

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0916. Min Act: -0.1700

Data Index: 777500 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0916. Min Act: -0.1700

Data Index: 462798 (C4 (Web Text))

Max Activating Token Index: 133

Text #16

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0758. Min Act: -0.1700

Data Index: 871695 (C4 (Web Text))

Max Activating Token Index: 548

Text #17

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0442. Min Act: -0.1700

Data Index: 620796 (C4 (Web Text))

Max Activating Token Index: 564

Text #18

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0442. Min Act: -0.1700

Data Index: 620877 (C4 (Web Text))

Max Activating Token Index: 378

Text #19

Max Range: 3.7193. Min Range: -3.7193

Max Act: 3.0600. Min Act: -0.1700

Data Index: 7978 (C4 (Web Text))

Max Activating Token Index: 438