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

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

Neuron 1727 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.5668. Min Act: -0.1700

Data Index: 1070920 (C4 (Web Text))

Max Activating Token Index: 120

Text #1

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.4106. Min Act: -0.1700

Data Index: 963795 (C4 (Web Text))

Max Activating Token Index: 728

Text #2

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.4106. Min Act: -0.1700

Data Index: 688322 (C4 (Web Text))

Max Activating Token Index: 812

Text #3

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.3793. Min Act: -0.1700

Data Index: 517493 (C4 (Web Text))

Max Activating Token Index: 284

Text #4

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.3481. Min Act: -0.1700

Data Index: 631317 (C4 (Web Text))

Max Activating Token Index: 198

Text #5

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.3168. Min Act: -0.1700

Data Index: 1351702 (C4 (Web Text))

Max Activating Token Index: 885

Text #6

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.3168. Min Act: -0.1700

Data Index: 706213 (C4 (Web Text))

Max Activating Token Index: 692

Text #7

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2856. Min Act: -0.1700

Data Index: 939484 (C4 (Web Text))

Max Activating Token Index: 331

Text #8

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 700286 (C4 (Web Text))

Max Activating Token Index: 503

Text #9

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 1207550 (C4 (Web Text))

Max Activating Token Index: 371

Text #10

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2231. Min Act: -0.1700

Data Index: 301764 (C4 (Web Text))

Max Activating Token Index: 673

Text #11

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 333859 (C4 (Web Text))

Max Activating Token Index: 265

Text #12

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 480352 (C4 (Web Text))

Max Activating Token Index: 751

Text #13

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2231. Min Act: -0.1700

Data Index: 1056432 (C4 (Web Text))

Max Activating Token Index: 30

Text #14

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.3168. Min Act: -0.1700

Data Index: 1002904 (C4 (Web Text))

Max Activating Token Index: 789

Text #15

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 274404 (C4 (Web Text))

Max Activating Token Index: 774

Text #16

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 348945 (C4 (Web Text))

Max Activating Token Index: 155

Text #17

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2856. Min Act: -0.1700

Data Index: 309641 (C4 (Web Text))

Max Activating Token Index: 694

Text #18

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.2543. Min Act: -0.1700

Data Index: 588359 (C4 (Web Text))

Max Activating Token Index: 832

Text #19

Max Range: 5.5668. Min Range: -5.5668

Max Act: 5.1918. Min Act: -0.1700

Data Index: 20956 (C4 (Web Text))

Max Activating Token Index: 375