Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1585 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8733. Min Act: -0.1700

Data Index: 902014 (C4 (Web Text))

Max Activating Token Index: 759

Text #1

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8889. Min Act: -0.1700

Data Index: 113567 (C4 (Web Text))

Max Activating Token Index: 281

Text #2

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8655. Min Act: -0.1700

Data Index: 156152 (C4 (Web Text))

Max Activating Token Index: 578

Text #3

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8610. Min Act: -0.1700

Data Index: 1450584 (Python Code)

Max Activating Token Index: 709

Text #4

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8437. Min Act: -0.1700

Data Index: 34980 (C4 (Web Text))

Max Activating Token Index: 627

Text #5

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8187. Min Act: -0.1700

Data Index: 596852 (C4 (Web Text))

Max Activating Token Index: 275

Text #6

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8192. Min Act: -0.1700

Data Index: 123330 (C4 (Web Text))

Max Activating Token Index: 851

Text #7

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7850. Min Act: -0.1700

Data Index: 1428476 (Python Code)

Max Activating Token Index: 519

Text #8

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7965. Min Act: -0.1700

Data Index: 961882 (C4 (Web Text))

Max Activating Token Index: 873

Text #9

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8022. Min Act: -0.1700

Data Index: 219538 (C4 (Web Text))

Max Activating Token Index: 827

Text #10

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.8060. Min Act: -0.1700

Data Index: 640810 (C4 (Web Text))

Max Activating Token Index: 265

Text #11

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7413. Min Act: -0.1700

Data Index: 1224096 (C4 (Web Text))

Max Activating Token Index: 941

Text #12

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7660. Min Act: -0.1700

Data Index: 1061351 (C4 (Web Text))

Max Activating Token Index: 583

Text #13

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7345. Min Act: -0.1700

Data Index: 720635 (C4 (Web Text))

Max Activating Token Index: 595

Text #14

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7425. Min Act: -0.1700

Data Index: 1391917 (Python Code)

Max Activating Token Index: 643

Text #15

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7297. Min Act: -0.1700

Data Index: 632932 (C4 (Web Text))

Max Activating Token Index: 36

Text #16

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7315. Min Act: -0.1700

Data Index: 61824 (C4 (Web Text))

Max Activating Token Index: 837

Text #17

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.7253. Min Act: -0.1700

Data Index: 698526 (C4 (Web Text))

Max Activating Token Index: 438

Text #18

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.6786. Min Act: -0.1700

Data Index: 415488 (C4 (Web Text))

Max Activating Token Index: 723

Text #19

Max Range: 3.8889. Min Range: -3.8889

Max Act: 3.6900. Min Act: -0.1700

Data Index: 236964 (C4 (Web Text))

Max Activating Token Index: 878