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

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

Neuron 658 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.7218. Min Act: -0.1700

Data Index: 1291972 (C4 (Web Text))

Max Activating Token Index: 385

Text #1

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.7843. Min Act: -0.1700

Data Index: 330744 (C4 (Web Text))

Max Activating Token Index: 841

Text #2

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.7843. Min Act: -0.1700

Data Index: 983758 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 395021 (C4 (Web Text))

Max Activating Token Index: 874

Text #4

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.8468. Min Act: -0.1700

Data Index: 1164160 (C4 (Web Text))

Max Activating Token Index: 406

Text #5

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.5968. Min Act: -0.1700

Data Index: 1088489 (C4 (Web Text))

Max Activating Token Index: 649

Text #6

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 185733 (C4 (Web Text))

Max Activating Token Index: 197

Text #7

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 532182 (C4 (Web Text))

Max Activating Token Index: 793

Text #8

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 170087 (C4 (Web Text))

Max Activating Token Index: 431

Text #9

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 1359711 (C4 (Web Text))

Max Activating Token Index: 205

Text #10

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6905. Min Act: -0.1700

Data Index: 744320 (C4 (Web Text))

Max Activating Token Index: 492

Text #11

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 425131 (C4 (Web Text))

Max Activating Token Index: 270

Text #12

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 454750 (C4 (Web Text))

Max Activating Token Index: 443

Text #13

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 1133873 (C4 (Web Text))

Max Activating Token Index: 498

Text #14

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.5655. Min Act: -0.1700

Data Index: 861144 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.4718. Min Act: -0.1700

Data Index: 1261190 (C4 (Web Text))

Max Activating Token Index: 490

Text #16

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.5968. Min Act: -0.1700

Data Index: 1208368 (C4 (Web Text))

Max Activating Token Index: 282

Text #17

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.5343. Min Act: -0.1700

Data Index: 562006 (C4 (Web Text))

Max Activating Token Index: 869

Text #18

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.6280. Min Act: -0.1700

Data Index: 711724 (C4 (Web Text))

Max Activating Token Index: 359

Text #19

Max Range: 6.8468. Min Range: -6.8468

Max Act: 6.5968. Min Act: -0.1700

Data Index: 665837 (C4 (Web Text))

Max Activating Token Index: 646