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

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

Neuron 145 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1889. Min Range: -5.1889

Max Act: 5.1577. Min Act: -0.1700

Data Index: 1089776 (C4 (Web Text))

Max Activating Token Index: 692

Text #1

Max Range: 5.1889. Min Range: -5.1889

Max Act: 5.1889. Min Act: -0.1700

Data Index: 659749 (C4 (Web Text))

Max Activating Token Index: 978

Text #2

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.9077. Min Act: -0.1700

Data Index: 800353 (C4 (Web Text))

Max Activating Token Index: 648

Text #3

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.7827. Min Act: -0.1700

Data Index: 927942 (C4 (Web Text))

Max Activating Token Index: 427

Text #4

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.8764. Min Act: -0.1700

Data Index: 456749 (C4 (Web Text))

Max Activating Token Index: 470

Text #5

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.8139. Min Act: -0.1700

Data Index: 1170364 (C4 (Web Text))

Max Activating Token Index: 472

Text #6

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.8764. Min Act: -0.1700

Data Index: 1964 (C4 (Web Text))

Max Activating Token Index: 531

Text #7

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.7827. Min Act: -0.1700

Data Index: 45886 (C4 (Web Text))

Max Activating Token Index: 108

Text #8

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.7514. Min Act: -0.1700

Data Index: 757130 (C4 (Web Text))

Max Activating Token Index: 272

Text #9

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6889. Min Act: -0.1700

Data Index: 31341 (C4 (Web Text))

Max Activating Token Index: 363

Text #10

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.7514. Min Act: -0.1700

Data Index: 977328 (C4 (Web Text))

Max Activating Token Index: 845

Text #11

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.7827. Min Act: -0.1699

Data Index: 1234939 (C4 (Web Text))

Max Activating Token Index: 804

Text #12

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6889. Min Act: -0.1700

Data Index: 832646 (C4 (Web Text))

Max Activating Token Index: 112

Text #13

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6576. Min Act: -0.1699

Data Index: 188963 (C4 (Web Text))

Max Activating Token Index: 957

Text #14

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6576. Min Act: -0.1700

Data Index: 875617 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6889. Min Act: -0.1700

Data Index: 6820 (C4 (Web Text))

Max Activating Token Index: 392

Text #16

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6889. Min Act: -0.1700

Data Index: 290255 (C4 (Web Text))

Max Activating Token Index: 921

Text #17

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.5639. Min Act: -0.1700

Data Index: 6909 (C4 (Web Text))

Max Activating Token Index: 813

Text #18

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6889. Min Act: -0.1700

Data Index: 932301 (C4 (Web Text))

Max Activating Token Index: 958

Text #19

Max Range: 5.1889. Min Range: -5.1889

Max Act: 4.6576. Min Act: -0.1700

Data Index: 932740 (C4 (Web Text))

Max Activating Token Index: 763