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

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

Neuron 244 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.3922. Min Act: -0.1700

Data Index: 148526 (C4 (Web Text))

Max Activating Token Index: 342

Text #1

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.4079. Min Act: -0.1700

Data Index: 1352554 (C4 (Web Text))

Max Activating Token Index: 826

Text #2

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.3451. Min Act: -0.1700

Data Index: 1291245 (C4 (Web Text))

Max Activating Token Index: 613

Text #3

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.3451. Min Act: -0.1700

Data Index: 651318 (C4 (Web Text))

Max Activating Token Index: 24

Text #4

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.3137. Min Act: -0.1700

Data Index: 427203 (C4 (Web Text))

Max Activating Token Index: 295

Text #5

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.3137. Min Act: -0.1700

Data Index: 457161 (C4 (Web Text))

Max Activating Token Index: 1010

Text #6

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2823. Min Act: -0.1700

Data Index: 533673 (C4 (Web Text))

Max Activating Token Index: 184

Text #7

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2823. Min Act: -0.1700

Data Index: 967017 (C4 (Web Text))

Max Activating Token Index: 470

Text #8

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2509. Min Act: -0.1700

Data Index: 459115 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2509. Min Act: -0.1700

Data Index: 842964 (C4 (Web Text))

Max Activating Token Index: 89

Text #10

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2509. Min Act: -0.1700

Data Index: 302756 (C4 (Web Text))

Max Activating Token Index: 14

Text #11

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2509. Min Act: -0.1700

Data Index: 1294983 (C4 (Web Text))

Max Activating Token Index: 310

Text #12

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2352. Min Act: -0.1700

Data Index: 270452 (C4 (Web Text))

Max Activating Token Index: 756

Text #13

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2195. Min Act: -0.1700

Data Index: 671791 (C4 (Web Text))

Max Activating Token Index: 752

Text #14

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2352. Min Act: -0.1700

Data Index: 88949 (C4 (Web Text))

Max Activating Token Index: 366

Text #15

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2509. Min Act: -0.1700

Data Index: 159186 (C4 (Web Text))

Max Activating Token Index: 73

Text #16

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2195. Min Act: -0.1700

Data Index: 861724 (C4 (Web Text))

Max Activating Token Index: 554

Text #17

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2352. Min Act: -0.1700

Data Index: 1253009 (C4 (Web Text))

Max Activating Token Index: 290

Text #18

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2195. Min Act: -0.1700

Data Index: 864942 (C4 (Web Text))

Max Activating Token Index: 996

Text #19

Max Range: 3.4079. Min Range: -3.4079

Max Act: 3.2352. Min Act: -0.1700

Data Index: 426484 (C4 (Web Text))

Max Activating Token Index: 598