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

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

Neuron 860 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.4272. Min Act: -0.1700

Data Index: 1102790 (C4 (Web Text))

Max Activating Token Index: 149

Text #1

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.3335. Min Act: -0.1699

Data Index: 408707 (C4 (Web Text))

Max Activating Token Index: 139

Text #2

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.2397. Min Act: -0.1700

Data Index: 509000 (C4 (Web Text))

Max Activating Token Index: 825

Text #3

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.1147. Min Act: -0.1700

Data Index: 1181011 (C4 (Web Text))

Max Activating Token Index: 403

Text #4

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.1147. Min Act: -0.1700

Data Index: 205767 (C4 (Web Text))

Max Activating Token Index: 789

Text #5

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.1772. Min Act: -0.1700

Data Index: 738357 (C4 (Web Text))

Max Activating Token Index: 66

Text #6

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0522. Min Act: -0.1700

Data Index: 968470 (C4 (Web Text))

Max Activating Token Index: 242

Text #7

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.1147. Min Act: -0.1700

Data Index: 873181 (C4 (Web Text))

Max Activating Token Index: 146

Text #8

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0835. Min Act: -0.1700

Data Index: 1228406 (C4 (Web Text))

Max Activating Token Index: 542

Text #9

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0522. Min Act: -0.1700

Data Index: 1222896 (C4 (Web Text))

Max Activating Token Index: 199

Text #10

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0210. Min Act: -0.1700

Data Index: 1026258 (C4 (Web Text))

Max Activating Token Index: 92

Text #11

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0210. Min Act: -0.1700

Data Index: 1262864 (C4 (Web Text))

Max Activating Token Index: 953

Text #12

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0835. Min Act: -0.1700

Data Index: 670220 (C4 (Web Text))

Max Activating Token Index: 438

Text #13

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0522. Min Act: -0.1700

Data Index: 786524 (C4 (Web Text))

Max Activating Token Index: 174

Text #14

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0522. Min Act: -0.1700

Data Index: 1359089 (C4 (Web Text))

Max Activating Token Index: 134

Text #15

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0522. Min Act: -0.1700

Data Index: 184070 (C4 (Web Text))

Max Activating Token Index: 921

Text #16

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0210. Min Act: -0.1699

Data Index: 1179719 (C4 (Web Text))

Max Activating Token Index: 555

Text #17

Max Range: 5.4272. Min Range: -5.4272

Max Act: 4.9897. Min Act: -0.1700

Data Index: 739936 (C4 (Web Text))

Max Activating Token Index: 71

Text #18

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0210. Min Act: -0.1700

Data Index: 1213706 (C4 (Web Text))

Max Activating Token Index: 478

Text #19

Max Range: 5.4272. Min Range: -5.4272

Max Act: 5.0210. Min Act: -0.1700

Data Index: 652494 (C4 (Web Text))

Max Activating Token Index: 483