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

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

Neuron 1384 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: 5.7812. Min Range: -5.7812

Max Act: 5.7812. Min Act: -0.1700

Data Index: 1582144 (Python Code)

Max Activating Token Index: 957

Text #1

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4375. Min Act: -0.1700

Data Index: 1459705 (Python Code)

Max Activating Token Index: 900

Text #2

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.5625. Min Act: -0.1700

Data Index: 177663 (C4 (Web Text))

Max Activating Token Index: 662

Text #3

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.2500. Min Act: -0.1700

Data Index: 1425160 (Python Code)

Max Activating Token Index: 1021

Text #4

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.5000. Min Act: -0.1700

Data Index: 1420056 (Python Code)

Max Activating Token Index: 966

Text #5

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.6250. Min Act: -0.1700

Data Index: 349854 (C4 (Web Text))

Max Activating Token Index: 796

Text #6

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4062. Min Act: -0.1700

Data Index: 1468407 (Python Code)

Max Activating Token Index: 1010

Text #7

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4375. Min Act: -0.1700

Data Index: 1396828 (Python Code)

Max Activating Token Index: 699

Text #8

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4687. Min Act: -0.1700

Data Index: 1561211 (Python Code)

Max Activating Token Index: 895

Text #9

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4687. Min Act: -0.1700

Data Index: 27507 (C4 (Web Text))

Max Activating Token Index: 699

Text #10

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4375. Min Act: -0.1700

Data Index: 67402 (C4 (Web Text))

Max Activating Token Index: 763

Text #11

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.2187. Min Act: -0.1700

Data Index: 771654 (C4 (Web Text))

Max Activating Token Index: 237

Text #12

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.2500. Min Act: -0.1700

Data Index: 1527951 (Python Code)

Max Activating Token Index: 953

Text #13

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.5937. Min Act: -0.1700

Data Index: 1445248 (Python Code)

Max Activating Token Index: 970

Text #14

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4062. Min Act: -0.1700

Data Index: 975208 (C4 (Web Text))

Max Activating Token Index: 697

Text #15

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4375. Min Act: -0.1700

Data Index: 1018720 (C4 (Web Text))

Max Activating Token Index: 796

Text #16

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.3437. Min Act: -0.1700

Data Index: 1101954 (C4 (Web Text))

Max Activating Token Index: 995

Text #17

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.2500. Min Act: -0.1700

Data Index: 1593103 (Python Code)

Max Activating Token Index: 602

Text #18

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.2500. Min Act: -0.1700

Data Index: 662611 (C4 (Web Text))

Max Activating Token Index: 425

Text #19

Max Range: 5.7812. Min Range: -5.7812

Max Act: 5.4062. Min Act: -0.1700

Data Index: 476132 (C4 (Web Text))

Max Activating Token Index: 32