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

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

Neuron 1946 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: 3.2675. Min Range: -3.2675

Max Act: 3.2675. Min Act: -0.1700

Data Index: 1464718 (Python Code)

Max Activating Token Index: 1012

Text #1

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1416. Min Act: -0.1700

Data Index: 270134 (C4 (Web Text))

Max Activating Token Index: 781

Text #2

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1258. Min Act: -0.1700

Data Index: 432639 (C4 (Web Text))

Max Activating Token Index: 880

Text #3

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1416. Min Act: -0.1700

Data Index: 759847 (C4 (Web Text))

Max Activating Token Index: 791

Text #4

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1731. Min Act: -0.1700

Data Index: 1216880 (C4 (Web Text))

Max Activating Token Index: 762

Text #5

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1101. Min Act: -0.1700

Data Index: 17705 (C4 (Web Text))

Max Activating Token Index: 742

Text #6

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1101. Min Act: -0.1700

Data Index: 1149143 (C4 (Web Text))

Max Activating Token Index: 901

Text #7

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1101. Min Act: -0.1700

Data Index: 20544 (C4 (Web Text))

Max Activating Token Index: 872

Text #8

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0628. Min Act: -0.1700

Data Index: 52826 (C4 (Web Text))

Max Activating Token Index: 704

Text #9

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0785. Min Act: -0.1700

Data Index: 237401 (C4 (Web Text))

Max Activating Token Index: 728

Text #10

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1101. Min Act: -0.1700

Data Index: 379110 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.1258. Min Act: -0.1700

Data Index: 124139 (C4 (Web Text))

Max Activating Token Index: 896

Text #12

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0470. Min Act: -0.1700

Data Index: 602634 (C4 (Web Text))

Max Activating Token Index: 995

Text #13

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0470. Min Act: -0.1700

Data Index: 1418848 (Python Code)

Max Activating Token Index: 417

Text #14

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0785. Min Act: -0.1700

Data Index: 414810 (C4 (Web Text))

Max Activating Token Index: 928

Text #15

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0628. Min Act: -0.1700

Data Index: 77011 (C4 (Web Text))

Max Activating Token Index: 894

Text #16

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0785. Min Act: -0.1700

Data Index: 448551 (C4 (Web Text))

Max Activating Token Index: 904

Text #17

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0470. Min Act: -0.1700

Data Index: 483065 (C4 (Web Text))

Max Activating Token Index: 1008

Text #18

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0154. Min Act: -0.1700

Data Index: 1302503 (C4 (Web Text))

Max Activating Token Index: 655

Text #19

Max Range: 3.2675. Min Range: -3.2675

Max Act: 3.0312. Min Act: -0.1700

Data Index: 699004 (C4 (Web Text))

Max Activating Token Index: 838