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

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

Neuron 717 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.6635. Min Act: -0.1700

Data Index: 1380560 (Python Code)

Max Activating Token Index: 234

Text #1

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.6009. Min Act: -0.1700

Data Index: 1396906 (Python Code)

Max Activating Token Index: 670

Text #2

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5695. Min Act: -0.1700

Data Index: 901547 (C4 (Web Text))

Max Activating Token Index: 679

Text #3

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5695. Min Act: -0.1700

Data Index: 1511824 (Python Code)

Max Activating Token Index: 645

Text #4

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5382. Min Act: -0.1700

Data Index: 692767 (C4 (Web Text))

Max Activating Token Index: 729

Text #5

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5382. Min Act: -0.1700

Data Index: 1548384 (Python Code)

Max Activating Token Index: 509

Text #6

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5382. Min Act: -0.1700

Data Index: 1385329 (Python Code)

Max Activating Token Index: 845

Text #7

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5069. Min Act: -0.1700

Data Index: 1561211 (Python Code)

Max Activating Token Index: 838

Text #8

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5069. Min Act: -0.1700

Data Index: 1506953 (Python Code)

Max Activating Token Index: 449

Text #9

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.5069. Min Act: -0.1700

Data Index: 1449666 (Python Code)

Max Activating Token Index: 660

Text #10

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4755. Min Act: -0.1700

Data Index: 1633090 (Python Code)

Max Activating Token Index: 579

Text #11

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 1544245 (Python Code)

Max Activating Token Index: 53

Text #12

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 734614 (C4 (Web Text))

Max Activating Token Index: 972

Text #13

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 544946 (C4 (Web Text))

Max Activating Token Index: 300

Text #14

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4755. Min Act: -0.1700

Data Index: 1471832 (Python Code)

Max Activating Token Index: 932

Text #15

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 1383669 (Python Code)

Max Activating Token Index: 459

Text #16

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 1375543 (Python Code)

Max Activating Token Index: 473

Text #17

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4285. Min Act: -0.1700

Data Index: 1458795 (Python Code)

Max Activating Token Index: 282

Text #18

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4285. Min Act: -0.1700

Data Index: 1589275 (Python Code)

Max Activating Token Index: 931

Text #19

Max Range: 3.6635. Min Range: -3.6635

Max Act: 3.4442. Min Act: -0.1700

Data Index: 1123138 (C4 (Web Text))

Max Activating Token Index: 871