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

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

Neuron 1915 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: 4.5441. Min Range: -4.5441

Max Act: 4.5441. Min Act: -0.1700

Data Index: 536222 (C4 (Web Text))

Max Activating Token Index: 864

Text #1

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.4480. Min Act: -0.1699

Data Index: 1133898 (C4 (Web Text))

Max Activating Token Index: 1005

Text #2

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.2718. Min Act: -0.1699

Data Index: 642394 (C4 (Web Text))

Max Activating Token Index: 895

Text #3

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.2525. Min Act: -0.1700

Data Index: 1240530 (C4 (Web Text))

Max Activating Token Index: 320

Text #4

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.1946. Min Act: -0.1699

Data Index: 1206959 (C4 (Web Text))

Max Activating Token Index: 797

Text #5

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.1748. Min Act: -0.1694

Data Index: 1343483 (C4 (Web Text))

Max Activating Token Index: 487

Text #6

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0969. Min Act: -0.1700

Data Index: 1325551 (C4 (Web Text))

Max Activating Token Index: 930

Text #7

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0844. Min Act: -0.1700

Data Index: 1361065 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.1081. Min Act: -0.1700

Data Index: 1525106 (Python Code)

Max Activating Token Index: 123

Text #9

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0844. Min Act: -0.1691

Data Index: 1375453 (Python Code)

Max Activating Token Index: 863

Text #10

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0508. Min Act: -0.1700

Data Index: 424204 (C4 (Web Text))

Max Activating Token Index: 657

Text #11

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0382. Min Act: -0.1699

Data Index: 854374 (C4 (Web Text))

Max Activating Token Index: 969

Text #12

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0377. Min Act: -0.1700

Data Index: 1376253 (Python Code)

Max Activating Token Index: 182

Text #13

Max Range: 4.5441. Min Range: -4.5441

Max Act: 4.0147. Min Act: -0.1699

Data Index: 541859 (C4 (Web Text))

Max Activating Token Index: 578

Text #14

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9738. Min Act: -0.1700

Data Index: 519757 (C4 (Web Text))

Max Activating Token Index: 1014

Text #15

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9825. Min Act: -0.1700

Data Index: 155302 (C4 (Web Text))

Max Activating Token Index: 563

Text #16

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9852. Min Act: -0.1700

Data Index: 839687 (C4 (Web Text))

Max Activating Token Index: 475

Text #17

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9879. Min Act: -0.1699

Data Index: 1490259 (Python Code)

Max Activating Token Index: 481

Text #18

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9645. Min Act: -0.1700

Data Index: 1623940 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 4.5441. Min Range: -4.5441

Max Act: 3.9369. Min Act: -0.1700

Data Index: 949617 (C4 (Web Text))

Max Activating Token Index: 664