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

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

Neuron 414 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: 5.7307. Min Range: -5.7307

Max Act: 5.7307. Min Act: -0.1699

Data Index: 763046 (C4 (Web Text))

Max Activating Token Index: 293

Text #1

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.6135. Min Act: -0.1699

Data Index: 810563 (C4 (Web Text))

Max Activating Token Index: 479

Text #2

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.3987. Min Act: -0.1699

Data Index: 626837 (C4 (Web Text))

Max Activating Token Index: 789

Text #3

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.3518. Min Act: -0.1699

Data Index: 1267111 (C4 (Web Text))

Max Activating Token Index: 886

Text #4

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.3206. Min Act: -0.1699

Data Index: 201117 (C4 (Web Text))

Max Activating Token Index: 877

Text #5

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.3127. Min Act: -0.1699

Data Index: 469290 (C4 (Web Text))

Max Activating Token Index: 436

Text #6

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.2541. Min Act: -0.1699

Data Index: 309317 (C4 (Web Text))

Max Activating Token Index: 342

Text #7

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.2776. Min Act: -0.1699

Data Index: 1352530 (C4 (Web Text))

Max Activating Token Index: 144

Text #8

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.2424. Min Act: -0.1699

Data Index: 775889 (C4 (Web Text))

Max Activating Token Index: 178

Text #9

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.2307. Min Act: -0.1699

Data Index: 926237 (C4 (Web Text))

Max Activating Token Index: 947

Text #10

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1643. Min Act: -0.1699

Data Index: 1476819 (Python Code)

Max Activating Token Index: 124

Text #11

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1331. Min Act: -0.1699

Data Index: 1506684 (Python Code)

Max Activating Token Index: 781

Text #12

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.2151. Min Act: -0.1699

Data Index: 1170711 (C4 (Web Text))

Max Activating Token Index: 609

Text #13

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1487. Min Act: -0.1699

Data Index: 523648 (C4 (Web Text))

Max Activating Token Index: 145

Text #14

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1916. Min Act: -0.1699

Data Index: 1177488 (C4 (Web Text))

Max Activating Token Index: 804

Text #15

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1760. Min Act: -0.1699

Data Index: 1073460 (C4 (Web Text))

Max Activating Token Index: 494

Text #16

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1057. Min Act: -0.1699

Data Index: 316639 (C4 (Web Text))

Max Activating Token Index: 549

Text #17

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1643. Min Act: -0.1699

Data Index: 1617323 (Python Code)

Max Activating Token Index: 295

Text #18

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.1174. Min Act: -0.1699

Data Index: 851189 (C4 (Web Text))

Max Activating Token Index: 636

Text #19

Max Range: 5.7307. Min Range: -5.7307

Max Act: 5.0745. Min Act: -0.1699

Data Index: 797183 (C4 (Web Text))

Max Activating Token Index: 51