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

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

Neuron 22 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: 7.8344. Min Range: -7.8344

Max Act: 7.8344. Min Act: -0.1700

Data Index: 1630257 (Python Code)

Max Activating Token Index: 736

Text #1

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.8031. Min Act: -0.1700

Data Index: 1545529 (Python Code)

Max Activating Token Index: 98

Text #2

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.6000. Min Act: -0.1700

Data Index: 1586479 (Python Code)

Max Activating Token Index: 287

Text #3

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.6156. Min Act: -0.1700

Data Index: 1646326 (Python Code)

Max Activating Token Index: 756

Text #4

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.5844. Min Act: -0.1700

Data Index: 1418099 (Python Code)

Max Activating Token Index: 815

Text #5

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.5375. Min Act: -0.1700

Data Index: 1495469 (Python Code)

Max Activating Token Index: 942

Text #6

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.5219. Min Act: -0.1700

Data Index: 1583459 (Python Code)

Max Activating Token Index: 500

Text #7

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.5063. Min Act: -0.1700

Data Index: 1571274 (Python Code)

Max Activating Token Index: 1018

Text #8

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.5063. Min Act: -0.1700

Data Index: 1545061 (Python Code)

Max Activating Token Index: 837

Text #9

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.4281. Min Act: -0.1700

Data Index: 1646372 (Python Code)

Max Activating Token Index: 620

Text #10

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.4594. Min Act: -0.1700

Data Index: 1482316 (Python Code)

Max Activating Token Index: 740

Text #11

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.4281. Min Act: -0.1700

Data Index: 1605879 (Python Code)

Max Activating Token Index: 957

Text #12

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3969. Min Act: -0.1700

Data Index: 1290842 (C4 (Web Text))

Max Activating Token Index: 1005

Text #13

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3500. Min Act: -0.1700

Data Index: 1126515 (C4 (Web Text))

Max Activating Token Index: 626

Text #14

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.4125. Min Act: -0.1700

Data Index: 1651576 (Python Code)

Max Activating Token Index: 1016

Text #15

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3656. Min Act: -0.1700

Data Index: 1445501 (Python Code)

Max Activating Token Index: 624

Text #16

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3188. Min Act: -0.1700

Data Index: 1543496 (Python Code)

Max Activating Token Index: 793

Text #17

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.2719. Min Act: -0.1700

Data Index: 1623413 (Python Code)

Max Activating Token Index: 409

Text #18

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3188. Min Act: -0.1700

Data Index: 1642831 (Python Code)

Max Activating Token Index: 1016

Text #19

Max Range: 7.8344. Min Range: -7.8344

Max Act: 7.3500. Min Act: -0.1700

Data Index: 1652819 (Python Code)

Max Activating Token Index: 699