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

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

Neuron 1395 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.1689. Min Act: -0.1700

Data Index: 969123 (C4 (Web Text))

Max Activating Token Index: 238

Text #1

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.1374. Min Act: -0.1700

Data Index: 1454183 (Python Code)

Max Activating Token Index: 709

Text #2

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0901. Min Act: -0.1700

Data Index: 339091 (C4 (Web Text))

Max Activating Token Index: 276

Text #3

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0901. Min Act: -0.1700

Data Index: 452039 (C4 (Web Text))

Max Activating Token Index: 657

Text #4

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0901. Min Act: -0.1700

Data Index: 1360523 (Python Code)

Max Activating Token Index: 101

Text #5

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0585. Min Act: -0.1700

Data Index: 74859 (C4 (Web Text))

Max Activating Token Index: 16

Text #6

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0428. Min Act: -0.1700

Data Index: 508677 (C4 (Web Text))

Max Activating Token Index: 880

Text #7

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0428. Min Act: -0.1700

Data Index: 1375824 (Python Code)

Max Activating Token Index: 626

Text #8

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 853987 (C4 (Web Text))

Max Activating Token Index: 64

Text #9

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 1548396 (Python Code)

Max Activating Token Index: 507

Text #10

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 526551 (C4 (Web Text))

Max Activating Token Index: 951

Text #11

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 528934 (C4 (Web Text))

Max Activating Token Index: 822

Text #12

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 122237 (C4 (Web Text))

Max Activating Token Index: 21

Text #13

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 984995 (C4 (Web Text))

Max Activating Token Index: 265

Text #14

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 696780 (C4 (Web Text))

Max Activating Token Index: 15

Text #15

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 890931 (C4 (Web Text))

Max Activating Token Index: 27

Text #16

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 815114 (C4 (Web Text))

Max Activating Token Index: 427

Text #17

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 874064 (C4 (Web Text))

Max Activating Token Index: 938

Text #18

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0270. Min Act: -0.1700

Data Index: 1501140 (Python Code)

Max Activating Token Index: 274

Text #19

Max Range: 3.1689. Min Range: -3.1689

Max Act: 3.0112. Min Act: -0.1700

Data Index: 1079028 (C4 (Web Text))

Max Activating Token Index: 525