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

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

Neuron 1027 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: 3.5851. Min Range: -3.5851

Max Act: 3.5224. Min Act: -0.1700

Data Index: 986962 (C4 (Web Text))

Max Activating Token Index: 832

Text #1

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.5851. Min Act: -0.1700

Data Index: 1062468 (C4 (Web Text))

Max Activating Token Index: 654

Text #2

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.5538. Min Act: -0.1700

Data Index: 844950 (C4 (Web Text))

Max Activating Token Index: 577

Text #3

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4911. Min Act: -0.1700

Data Index: 994967 (C4 (Web Text))

Max Activating Token Index: 952

Text #4

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.5224. Min Act: -0.1700

Data Index: 634485 (C4 (Web Text))

Max Activating Token Index: 837

Text #5

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.5538. Min Act: -0.1700

Data Index: 649093 (C4 (Web Text))

Max Activating Token Index: 192

Text #6

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4598. Min Act: -0.1700

Data Index: 426648 (C4 (Web Text))

Max Activating Token Index: 820

Text #7

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4598. Min Act: -0.1700

Data Index: 963127 (C4 (Web Text))

Max Activating Token Index: 334

Text #8

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4598. Min Act: -0.1700

Data Index: 384208 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4911. Min Act: -0.1700

Data Index: 477896 (C4 (Web Text))

Max Activating Token Index: 100

Text #10

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4598. Min Act: -0.1700

Data Index: 54620 (C4 (Web Text))

Max Activating Token Index: 240

Text #11

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4441. Min Act: -0.1700

Data Index: 414433 (C4 (Web Text))

Max Activating Token Index: 699

Text #12

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4911. Min Act: -0.1700

Data Index: 1018515 (C4 (Web Text))

Max Activating Token Index: 977

Text #13

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.3970. Min Act: -0.1700

Data Index: 846623 (C4 (Web Text))

Max Activating Token Index: 961

Text #14

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4284. Min Act: -0.1700

Data Index: 470760 (C4 (Web Text))

Max Activating Token Index: 698

Text #15

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4284. Min Act: -0.1700

Data Index: 1151869 (C4 (Web Text))

Max Activating Token Index: 556

Text #16

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4284. Min Act: -0.1700

Data Index: 338764 (C4 (Web Text))

Max Activating Token Index: 438

Text #17

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4127. Min Act: -0.1700

Data Index: 261945 (C4 (Web Text))

Max Activating Token Index: 657

Text #18

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.3814. Min Act: -0.1700

Data Index: 522157 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 3.5851. Min Range: -3.5851

Max Act: 3.4127. Min Act: -0.1700

Data Index: 412174 (C4 (Web Text))

Max Activating Token Index: 898