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

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

Neuron 1300 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6874. Min Act: -0.1700

Data Index: 1329943 (C4 (Web Text))

Max Activating Token Index: 966

Text #1

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6404. Min Act: -0.1700

Data Index: 986438 (C4 (Web Text))

Max Activating Token Index: 658

Text #2

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6404. Min Act: -0.1700

Data Index: 211468 (C4 (Web Text))

Max Activating Token Index: 285

Text #3

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6091. Min Act: -0.1700

Data Index: 91580 (C4 (Web Text))

Max Activating Token Index: 654

Text #4

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6248. Min Act: -0.1700

Data Index: 1096175 (C4 (Web Text))

Max Activating Token Index: 615

Text #5

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6091. Min Act: -0.1700

Data Index: 1268466 (C4 (Web Text))

Max Activating Token Index: 575

Text #6

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.6091. Min Act: -0.1700

Data Index: 949850 (C4 (Web Text))

Max Activating Token Index: 658

Text #7

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5778. Min Act: -0.1700

Data Index: 265402 (C4 (Web Text))

Max Activating Token Index: 826

Text #8

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5308. Min Act: -0.1700

Data Index: 273084 (C4 (Web Text))

Max Activating Token Index: 178

Text #9

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5621. Min Act: -0.1700

Data Index: 1309678 (C4 (Web Text))

Max Activating Token Index: 157

Text #10

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5621. Min Act: -0.1700

Data Index: 453055 (C4 (Web Text))

Max Activating Token Index: 867

Text #11

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5621. Min Act: -0.1700

Data Index: 206841 (C4 (Web Text))

Max Activating Token Index: 538

Text #12

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5465. Min Act: -0.1700

Data Index: 98268 (C4 (Web Text))

Max Activating Token Index: 521

Text #13

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5465. Min Act: -0.1700

Data Index: 1052912 (C4 (Web Text))

Max Activating Token Index: 64

Text #14

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5621. Min Act: -0.1700

Data Index: 967144 (C4 (Web Text))

Max Activating Token Index: 341

Text #15

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5308. Min Act: -0.1700

Data Index: 521649 (C4 (Web Text))

Max Activating Token Index: 690

Text #16

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5308. Min Act: -0.1700

Data Index: 293815 (C4 (Web Text))

Max Activating Token Index: 199

Text #17

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5308. Min Act: -0.1700

Data Index: 368349 (C4 (Web Text))

Max Activating Token Index: 239

Text #18

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5151. Min Act: -0.1700

Data Index: 1056245 (C4 (Web Text))

Max Activating Token Index: 692

Text #19

Max Range: 3.6874. Min Range: -3.6874

Max Act: 3.5308. Min Act: -0.1700

Data Index: 952633 (C4 (Web Text))

Max Activating Token Index: 693