Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1960 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.7869. Min Act: -0.1700

Data Index: 304408 (C4 (Web Text))

Max Activating Token Index: 376

Text #1

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.7243. Min Act: -0.1700

Data Index: 824483 (C4 (Web Text))

Max Activating Token Index: 341

Text #2

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.7243. Min Act: -0.1700

Data Index: 892975 (C4 (Web Text))

Max Activating Token Index: 909

Text #3

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.6930. Min Act: -0.1700

Data Index: 435199 (C4 (Web Text))

Max Activating Token Index: 785

Text #4

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.6930. Min Act: -0.1700

Data Index: 1291057 (C4 (Web Text))

Max Activating Token Index: 270

Text #5

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5991. Min Act: -0.1700

Data Index: 1068344 (C4 (Web Text))

Max Activating Token Index: 51

Text #6

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5991. Min Act: -0.1700

Data Index: 1220717 (C4 (Web Text))

Max Activating Token Index: 398

Text #7

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5991. Min Act: -0.1700

Data Index: 425651 (C4 (Web Text))

Max Activating Token Index: 958

Text #8

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5678. Min Act: -0.1700

Data Index: 1337211 (C4 (Web Text))

Max Activating Token Index: 716

Text #9

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5991. Min Act: -0.1700

Data Index: 317059 (C4 (Web Text))

Max Activating Token Index: 961

Text #10

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5678. Min Act: -0.1700

Data Index: 768165 (C4 (Web Text))

Max Activating Token Index: 1010

Text #11

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5678. Min Act: -0.1700

Data Index: 1152916 (C4 (Web Text))

Max Activating Token Index: 95

Text #12

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5991. Min Act: -0.1700

Data Index: 84193 (C4 (Web Text))

Max Activating Token Index: 500

Text #13

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5365. Min Act: -0.1700

Data Index: 469526 (C4 (Web Text))

Max Activating Token Index: 338

Text #14

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5365. Min Act: -0.1700

Data Index: 11966 (C4 (Web Text))

Max Activating Token Index: 243

Text #15

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5365. Min Act: -0.1700

Data Index: 921917 (C4 (Web Text))

Max Activating Token Index: 949

Text #16

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5365. Min Act: -0.1700

Data Index: 186160 (C4 (Web Text))

Max Activating Token Index: 293

Text #17

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5678. Min Act: -0.1700

Data Index: 68440 (C4 (Web Text))

Max Activating Token Index: 176

Text #18

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.4738. Min Act: -0.1700

Data Index: 1372075 (Python Code)

Max Activating Token Index: 693

Text #19

Max Range: 3.7869. Min Range: -3.7869

Max Act: 3.5051. Min Act: -0.1700

Data Index: 1448219 (Python Code)

Max Activating Token Index: 633