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

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

Neuron 1810 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.6656. Min Act: -0.1700

Data Index: 722665 (C4 (Web Text))

Max Activating Token Index: 970

Text #1

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.6343. Min Act: -0.1700

Data Index: 203472 (C4 (Web Text))

Max Activating Token Index: 57

Text #2

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.6031. Min Act: -0.1700

Data Index: 1651474 (Python Code)

Max Activating Token Index: 42

Text #3

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.5093. Min Act: -0.1700

Data Index: 891643 (C4 (Web Text))

Max Activating Token Index: 768

Text #4

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.5718. Min Act: -0.1700

Data Index: 703753 (C4 (Web Text))

Max Activating Token Index: 55

Text #5

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.6343. Min Act: -0.1700

Data Index: 1656911 (Python Code)

Max Activating Token Index: 682

Text #6

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.5406. Min Act: -0.1700

Data Index: 928532 (C4 (Web Text))

Max Activating Token Index: 378

Text #7

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.4468. Min Act: -0.1700

Data Index: 1579450 (Python Code)

Max Activating Token Index: 39

Text #8

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.3218. Min Act: -0.1700

Data Index: 1523707 (Python Code)

Max Activating Token Index: 458

Text #9

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.3843. Min Act: -0.1700

Data Index: 134280 (C4 (Web Text))

Max Activating Token Index: 803

Text #10

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.3218. Min Act: -0.1700

Data Index: 1488473 (Python Code)

Max Activating Token Index: 567

Text #11

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.4468. Min Act: -0.1700

Data Index: 1585340 (Python Code)

Max Activating Token Index: 749

Text #12

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.4156. Min Act: -0.1700

Data Index: 146864 (C4 (Web Text))

Max Activating Token Index: 785

Text #13

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.3218. Min Act: -0.1700

Data Index: 1440642 (Python Code)

Max Activating Token Index: 122

Text #14

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.3531. Min Act: -0.1700

Data Index: 1518236 (Python Code)

Max Activating Token Index: 276

Text #15

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.2593. Min Act: -0.1700

Data Index: 1395814 (Python Code)

Max Activating Token Index: 722

Text #16

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.2593. Min Act: -0.1700

Data Index: 1497547 (Python Code)

Max Activating Token Index: 58

Text #17

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.2906. Min Act: -0.1700

Data Index: 1563583 (Python Code)

Max Activating Token Index: 488

Text #18

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.2593. Min Act: -0.1700

Data Index: 1519232 (Python Code)

Max Activating Token Index: 37

Text #19

Max Range: 5.6656. Min Range: -5.6656

Max Act: 5.2281. Min Act: -0.1700

Data Index: 1403243 (Python Code)

Max Activating Token Index: 878