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

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

Neuron 2016 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2975. Min Act: -0.1700

Data Index: 439861 (C4 (Web Text))

Max Activating Token Index: 902

Text #1

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2503. Min Act: -0.1700

Data Index: 1102529 (C4 (Web Text))

Max Activating Token Index: 33

Text #2

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2660. Min Act: -0.1700

Data Index: 595710 (C4 (Web Text))

Max Activating Token Index: 913

Text #3

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2503. Min Act: -0.1700

Data Index: 1318433 (C4 (Web Text))

Max Activating Token Index: 201

Text #4

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2189. Min Act: -0.1700

Data Index: 552241 (C4 (Web Text))

Max Activating Token Index: 422

Text #5

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.2346. Min Act: -0.1700

Data Index: 1348112 (C4 (Web Text))

Max Activating Token Index: 625

Text #6

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1717. Min Act: -0.1700

Data Index: 688822 (C4 (Web Text))

Max Activating Token Index: 31

Text #7

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1402. Min Act: -0.1700

Data Index: 1321527 (C4 (Web Text))

Max Activating Token Index: 69

Text #8

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1402. Min Act: -0.1700

Data Index: 698578 (C4 (Web Text))

Max Activating Token Index: 183

Text #9

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0929. Min Act: -0.1700

Data Index: 1066543 (C4 (Web Text))

Max Activating Token Index: 597

Text #10

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1244. Min Act: -0.1700

Data Index: 427610 (C4 (Web Text))

Max Activating Token Index: 88

Text #11

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1244. Min Act: -0.1700

Data Index: 608484 (C4 (Web Text))

Max Activating Token Index: 68

Text #12

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0929. Min Act: -0.1700

Data Index: 1239211 (C4 (Web Text))

Max Activating Token Index: 219

Text #13

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0929. Min Act: -0.1700

Data Index: 1162715 (C4 (Web Text))

Max Activating Token Index: 892

Text #14

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0771. Min Act: -0.1700

Data Index: 188314 (C4 (Web Text))

Max Activating Token Index: 802

Text #15

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1087. Min Act: -0.1700

Data Index: 1282699 (C4 (Web Text))

Max Activating Token Index: 96

Text #16

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1087. Min Act: -0.1700

Data Index: 150331 (C4 (Web Text))

Max Activating Token Index: 888

Text #17

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0929. Min Act: -0.1700

Data Index: 1281058 (C4 (Web Text))

Max Activating Token Index: 972

Text #18

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.0929. Min Act: -0.1700

Data Index: 1175541 (C4 (Web Text))

Max Activating Token Index: 246

Text #19

Max Range: 3.2975. Min Range: -3.2975

Max Act: 3.1087. Min Act: -0.1700

Data Index: 539163 (C4 (Web Text))

Max Activating Token Index: 787