Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 432 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.1472. Min Range: -5.1472

Max Act: 5.1472. Min Act: -0.1700

Data Index: 487597 (C4 (Web Text))

Max Activating Token Index: 926

Text #1

Max Range: 5.1472. Min Range: -5.1472

Max Act: 5.0847. Min Act: -0.1700

Data Index: 400674 (C4 (Web Text))

Max Activating Token Index: 607

Text #2

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.9597. Min Act: -0.1700

Data Index: 168718 (C4 (Web Text))

Max Activating Token Index: 81

Text #3

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.8972. Min Act: -0.1700

Data Index: 1599508 (Python Code)

Max Activating Token Index: 668

Text #4

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.8659. Min Act: -0.1700

Data Index: 301078 (C4 (Web Text))

Max Activating Token Index: 253

Text #5

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.9284. Min Act: -0.1700

Data Index: 825769 (C4 (Web Text))

Max Activating Token Index: 934

Text #6

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.8034. Min Act: -0.1700

Data Index: 838132 (C4 (Web Text))

Max Activating Token Index: 701

Text #7

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.8034. Min Act: -0.1700

Data Index: 773956 (C4 (Web Text))

Max Activating Token Index: 949

Text #8

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7722. Min Act: -0.1700

Data Index: 1650472 (Python Code)

Max Activating Token Index: 336

Text #9

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7409. Min Act: -0.1700

Data Index: 1459803 (Python Code)

Max Activating Token Index: 963

Text #10

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7097. Min Act: -0.1700

Data Index: 607643 (C4 (Web Text))

Max Activating Token Index: 506

Text #11

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7722. Min Act: -0.1700

Data Index: 725795 (C4 (Web Text))

Max Activating Token Index: 693

Text #12

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7409. Min Act: -0.1700

Data Index: 1128916 (C4 (Web Text))

Max Activating Token Index: 617

Text #13

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7409. Min Act: -0.1700

Data Index: 102789 (C4 (Web Text))

Max Activating Token Index: 764

Text #14

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.7409. Min Act: -0.1700

Data Index: 324495 (C4 (Web Text))

Max Activating Token Index: 1020

Text #15

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.6784. Min Act: -0.1700

Data Index: 1051715 (C4 (Web Text))

Max Activating Token Index: 973

Text #16

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.6472. Min Act: -0.1700

Data Index: 1129285 (C4 (Web Text))

Max Activating Token Index: 249

Text #17

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.6472. Min Act: -0.1700

Data Index: 427470 (C4 (Web Text))

Max Activating Token Index: 438

Text #18

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.6472. Min Act: -0.1700

Data Index: 1044980 (C4 (Web Text))

Max Activating Token Index: 982

Text #19

Max Range: 5.1472. Min Range: -5.1472

Max Act: 4.6472. Min Act: -0.1700

Data Index: 273015 (C4 (Web Text))

Max Activating Token Index: 439