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

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

Neuron 540 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.9963. Min Act: -0.1700

Data Index: 745209 (C4 (Web Text))

Max Activating Token Index: 694

Text #1

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8712. Min Act: -0.1700

Data Index: 371048 (C4 (Web Text))

Max Activating Token Index: 884

Text #2

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.9025. Min Act: -0.1700

Data Index: 437716 (C4 (Web Text))

Max Activating Token Index: 624

Text #3

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.9025. Min Act: -0.1700

Data Index: 1336611 (C4 (Web Text))

Max Activating Token Index: 610

Text #4

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8712. Min Act: -0.1700

Data Index: 1193524 (C4 (Web Text))

Max Activating Token Index: 712

Text #5

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8399. Min Act: -0.1700

Data Index: 98301 (C4 (Web Text))

Max Activating Token Index: 619

Text #6

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8399. Min Act: -0.1700

Data Index: 270982 (C4 (Web Text))

Max Activating Token Index: 693

Text #7

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 138227 (C4 (Web Text))

Max Activating Token Index: 657

Text #8

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8399. Min Act: -0.1700

Data Index: 1019505 (C4 (Web Text))

Max Activating Token Index: 814

Text #9

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.8087. Min Act: -0.1700

Data Index: 762402 (C4 (Web Text))

Max Activating Token Index: 775

Text #10

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 1036431 (C4 (Web Text))

Max Activating Token Index: 490

Text #11

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 175475 (C4 (Web Text))

Max Activating Token Index: 55

Text #12

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7461. Min Act: -0.1700

Data Index: 560314 (C4 (Web Text))

Max Activating Token Index: 930

Text #13

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7148. Min Act: -0.1700

Data Index: 412143 (C4 (Web Text))

Max Activating Token Index: 402

Text #14

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 1307379 (C4 (Web Text))

Max Activating Token Index: 468

Text #15

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 832822 (C4 (Web Text))

Max Activating Token Index: 603

Text #16

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7774. Min Act: -0.1700

Data Index: 547715 (C4 (Web Text))

Max Activating Token Index: 353

Text #17

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7148. Min Act: -0.1700

Data Index: 1136776 (C4 (Web Text))

Max Activating Token Index: 828

Text #18

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7148. Min Act: -0.1700

Data Index: 853638 (C4 (Web Text))

Max Activating Token Index: 613

Text #19

Max Range: 3.9963. Min Range: -3.9963

Max Act: 3.7148. Min Act: -0.1700

Data Index: 530421 (C4 (Web Text))

Max Activating Token Index: 467