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

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

Neuron 170 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: 5.3315. Min Range: -5.3315

Max Act: 5.3315. Min Act: -0.1700

Data Index: 365663 (C4 (Web Text))

Max Activating Token Index: 831

Text #1

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.1753. Min Act: -0.1700

Data Index: 808833 (C4 (Web Text))

Max Activating Token Index: 88

Text #2

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.1753. Min Act: -0.1700

Data Index: 350803 (C4 (Web Text))

Max Activating Token Index: 352

Text #3

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.1440. Min Act: -0.1700

Data Index: 1304768 (C4 (Web Text))

Max Activating Token Index: 719

Text #4

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.1128. Min Act: -0.1700

Data Index: 239062 (C4 (Web Text))

Max Activating Token Index: 909

Text #5

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.0190. Min Act: -0.1700

Data Index: 472668 (C4 (Web Text))

Max Activating Token Index: 432

Text #6

Max Range: 5.3315. Min Range: -5.3315

Max Act: 5.0190. Min Act: -0.1700

Data Index: 468802 (C4 (Web Text))

Max Activating Token Index: 830

Text #7

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9253. Min Act: -0.1700

Data Index: 693121 (C4 (Web Text))

Max Activating Token Index: 591

Text #8

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9253. Min Act: -0.1700

Data Index: 1096616 (C4 (Web Text))

Max Activating Token Index: 338

Text #9

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9565. Min Act: -0.1700

Data Index: 370937 (C4 (Web Text))

Max Activating Token Index: 502

Text #10

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9565. Min Act: -0.1700

Data Index: 666546 (C4 (Web Text))

Max Activating Token Index: 993

Text #11

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9565. Min Act: -0.1700

Data Index: 902566 (C4 (Web Text))

Max Activating Token Index: 156

Text #12

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9565. Min Act: -0.1700

Data Index: 1107350 (C4 (Web Text))

Max Activating Token Index: 942

Text #13

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9253. Min Act: -0.1700

Data Index: 241426 (C4 (Web Text))

Max Activating Token Index: 50

Text #14

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.9565. Min Act: -0.1700

Data Index: 714518 (C4 (Web Text))

Max Activating Token Index: 431

Text #15

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.7690. Min Act: -0.1700

Data Index: 149681 (C4 (Web Text))

Max Activating Token Index: 958

Text #16

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.8003. Min Act: -0.1700

Data Index: 777695 (C4 (Web Text))

Max Activating Token Index: 860

Text #17

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.8315. Min Act: -0.1700

Data Index: 417989 (C4 (Web Text))

Max Activating Token Index: 826

Text #18

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.8003. Min Act: -0.1700

Data Index: 488241 (C4 (Web Text))

Max Activating Token Index: 499

Text #19

Max Range: 5.3315. Min Range: -5.3315

Max Act: 4.8940. Min Act: -0.1700

Data Index: 843273 (C4 (Web Text))

Max Activating Token Index: 596