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

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

Neuron 1944 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: 5.5767. Min Range: -5.5767

Max Act: 5.5454. Min Act: -0.1700

Data Index: 234003 (C4 (Web Text))

Max Activating Token Index: 744

Text #1

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.5767. Min Act: -0.1700

Data Index: 294040 (C4 (Web Text))

Max Activating Token Index: 953

Text #2

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.4829. Min Act: -0.1700

Data Index: 403320 (C4 (Web Text))

Max Activating Token Index: 277

Text #3

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2954. Min Act: -0.1700

Data Index: 727577 (C4 (Web Text))

Max Activating Token Index: 989

Text #4

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.3579. Min Act: -0.1700

Data Index: 45535 (C4 (Web Text))

Max Activating Token Index: 423

Text #5

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.3267. Min Act: -0.1700

Data Index: 1006468 (C4 (Web Text))

Max Activating Token Index: 494

Text #6

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.3892. Min Act: -0.1700

Data Index: 253578 (C4 (Web Text))

Max Activating Token Index: 1022

Text #7

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 1128120 (C4 (Web Text))

Max Activating Token Index: 249

Text #8

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2329. Min Act: -0.1700

Data Index: 487711 (C4 (Web Text))

Max Activating Token Index: 294

Text #9

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 755042 (C4 (Web Text))

Max Activating Token Index: 497

Text #10

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 270813 (C4 (Web Text))

Max Activating Token Index: 323

Text #11

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2017. Min Act: -0.1700

Data Index: 119032 (C4 (Web Text))

Max Activating Token Index: 427

Text #12

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2329. Min Act: -0.1700

Data Index: 1015684 (C4 (Web Text))

Max Activating Token Index: 338

Text #13

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.1704. Min Act: -0.1700

Data Index: 1010594 (C4 (Web Text))

Max Activating Token Index: 862

Text #14

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1699

Data Index: 284490 (C4 (Web Text))

Max Activating Token Index: 44

Text #15

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 93325 (C4 (Web Text))

Max Activating Token Index: 276

Text #16

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2017. Min Act: -0.1700

Data Index: 855057 (C4 (Web Text))

Max Activating Token Index: 759

Text #17

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 722852 (C4 (Web Text))

Max Activating Token Index: 31

Text #18

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.2642. Min Act: -0.1700

Data Index: 190487 (C4 (Web Text))

Max Activating Token Index: 939

Text #19

Max Range: 5.5767. Min Range: -5.5767

Max Act: 5.1704. Min Act: -0.1700

Data Index: 16302 (C4 (Web Text))

Max Activating Token Index: 31