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

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

Neuron 1681 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: 5.6717. Min Range: -5.6717

Max Act: 5.6717. Min Act: -0.1700

Data Index: 734751 (C4 (Web Text))

Max Activating Token Index: 757

Text #1

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.3592. Min Act: -0.1700

Data Index: 1275599 (C4 (Web Text))

Max Activating Token Index: 451

Text #2

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.3592. Min Act: -0.1699

Data Index: 590960 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.3280. Min Act: -0.1700

Data Index: 867651 (C4 (Web Text))

Max Activating Token Index: 589

Text #4

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2655. Min Act: -0.1700

Data Index: 956687 (C4 (Web Text))

Max Activating Token Index: 1008

Text #5

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2967. Min Act: -0.1700

Data Index: 784354 (C4 (Web Text))

Max Activating Token Index: 663

Text #6

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2967. Min Act: -0.1700

Data Index: 596854 (C4 (Web Text))

Max Activating Token Index: 55

Text #7

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2967. Min Act: -0.1700

Data Index: 759189 (C4 (Web Text))

Max Activating Token Index: 450

Text #8

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2655. Min Act: -0.1700

Data Index: 261937 (C4 (Web Text))

Max Activating Token Index: 742

Text #9

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2967. Min Act: -0.1700

Data Index: 542020 (C4 (Web Text))

Max Activating Token Index: 483

Text #10

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2342. Min Act: -0.1700

Data Index: 532918 (C4 (Web Text))

Max Activating Token Index: 159

Text #11

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2655. Min Act: -0.1700

Data Index: 1049261 (C4 (Web Text))

Max Activating Token Index: 63

Text #12

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2030. Min Act: -0.1700

Data Index: 985489 (C4 (Web Text))

Max Activating Token Index: 861

Text #13

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2030. Min Act: -0.1700

Data Index: 81636 (C4 (Web Text))

Max Activating Token Index: 129

Text #14

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2655. Min Act: -0.1700

Data Index: 312249 (C4 (Web Text))

Max Activating Token Index: 728

Text #15

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.1717. Min Act: -0.1700

Data Index: 1326250 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2030. Min Act: -0.1700

Data Index: 47860 (C4 (Web Text))

Max Activating Token Index: 381

Text #17

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.1717. Min Act: -0.1700

Data Index: 995097 (C4 (Web Text))

Max Activating Token Index: 717

Text #18

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.1405. Min Act: -0.1700

Data Index: 1064732 (C4 (Web Text))

Max Activating Token Index: 602

Text #19

Max Range: 5.6717. Min Range: -5.6717

Max Act: 5.2967. Min Act: -0.1700

Data Index: 162539 (C4 (Web Text))

Max Activating Token Index: 363