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

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

Neuron 1886 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: 8.7569. Min Range: -8.7569

Max Act: 8.7569. Min Act: -0.1700

Data Index: 409167 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 8.7569. Min Range: -8.7569

Max Act: 8.0069. Min Act: -0.1700

Data Index: 812589 (C4 (Web Text))

Max Activating Token Index: 854

Text #2

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.8507. Min Act: -0.1700

Data Index: 1511424 (Python Code)

Max Activating Token Index: 359

Text #3

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.8507. Min Act: -0.1700

Data Index: 1558281 (Python Code)

Max Activating Token Index: 885

Text #4

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.8507. Min Act: -0.1700

Data Index: 1292730 (C4 (Web Text))

Max Activating Token Index: 712

Text #5

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.8194. Min Act: -0.1700

Data Index: 1487125 (Python Code)

Max Activating Token Index: 849

Text #6

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6632. Min Act: -0.1700

Data Index: 582382 (C4 (Web Text))

Max Activating Token Index: 375

Text #7

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6632. Min Act: -0.1700

Data Index: 1609049 (Python Code)

Max Activating Token Index: 885

Text #8

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6632. Min Act: -0.1700

Data Index: 414153 (C4 (Web Text))

Max Activating Token Index: 367

Text #9

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6319. Min Act: -0.1700

Data Index: 1397005 (Python Code)

Max Activating Token Index: 502

Text #10

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6319. Min Act: -0.1700

Data Index: 1343077 (C4 (Web Text))

Max Activating Token Index: 165

Text #11

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6007. Min Act: -0.1700

Data Index: 1527645 (Python Code)

Max Activating Token Index: 628

Text #12

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.5382. Min Act: -0.1700

Data Index: 1562960 (Python Code)

Max Activating Token Index: 717

Text #13

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.6007. Min Act: -0.1700

Data Index: 821784 (C4 (Web Text))

Max Activating Token Index: 356

Text #14

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.5694. Min Act: -0.1700

Data Index: 741085 (C4 (Web Text))

Max Activating Token Index: 693

Text #15

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.5069. Min Act: -0.1700

Data Index: 1624975 (Python Code)

Max Activating Token Index: 733

Text #16

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.5382. Min Act: -0.1700

Data Index: 1489513 (Python Code)

Max Activating Token Index: 481

Text #17

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.4757. Min Act: -0.1700

Data Index: 1604676 (Python Code)

Max Activating Token Index: 805

Text #18

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.4132. Min Act: -0.1700

Data Index: 1583646 (Python Code)

Max Activating Token Index: 936

Text #19

Max Range: 8.7569. Min Range: -8.7569

Max Act: 7.5694. Min Act: -0.1700

Data Index: 1556508 (Python Code)

Max Activating Token Index: 881