Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 539 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.9220. Min Act: -0.1683

Data Index: 1430663 (Python Code)

Max Activating Token Index: 519

Text #1

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.7030. Min Act: -0.1700

Data Index: 131553 (C4 (Web Text))

Max Activating Token Index: 545

Text #2

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6717. Min Act: -0.1700

Data Index: 1288394 (C4 (Web Text))

Max Activating Token Index: 290

Text #3

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6717. Min Act: -0.1662

Data Index: 1592816 (Python Code)

Max Activating Token Index: 226

Text #4

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6874. Min Act: -0.1700

Data Index: 1536884 (Python Code)

Max Activating Token Index: 21

Text #5

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6874. Min Act: -0.1699

Data Index: 1146531 (C4 (Web Text))

Max Activating Token Index: 411

Text #6

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6717. Min Act: -0.1700

Data Index: 1102207 (C4 (Web Text))

Max Activating Token Index: 526

Text #7

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6091. Min Act: -0.1700

Data Index: 635170 (C4 (Web Text))

Max Activating Token Index: 210

Text #8

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.6404. Min Act: -0.1700

Data Index: 1263504 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5935. Min Act: -0.1700

Data Index: 915451 (C4 (Web Text))

Max Activating Token Index: 589

Text #10

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5621. Min Act: -0.1700

Data Index: 757413 (C4 (Web Text))

Max Activating Token Index: 44

Text #11

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5465. Min Act: -0.1700

Data Index: 1559657 (Python Code)

Max Activating Token Index: 15

Text #12

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5621. Min Act: -0.1700

Data Index: 136912 (C4 (Web Text))

Max Activating Token Index: 23

Text #13

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5465. Min Act: -0.1700

Data Index: 1324102 (C4 (Web Text))

Max Activating Token Index: 356

Text #14

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.5151. Min Act: -0.1700

Data Index: 488296 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.4995. Min Act: -0.1700

Data Index: 41347 (C4 (Web Text))

Max Activating Token Index: 335

Text #16

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.4681. Min Act: -0.1700

Data Index: 1364076 (Python Code)

Max Activating Token Index: 603

Text #17

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.4838. Min Act: -0.1700

Data Index: 1243634 (C4 (Web Text))

Max Activating Token Index: 240

Text #18

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.4995. Min Act: -0.1700

Data Index: 163747 (C4 (Web Text))

Max Activating Token Index: 685

Text #19

Max Range: 3.9220. Min Range: -3.9220

Max Act: 3.4838. Min Act: -0.1699

Data Index: 740375 (C4 (Web Text))

Max Activating Token Index: 226