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

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

Neuron 1504 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1065. Min Range: -5.1065

Max Act: 5.1065. Min Act: -0.1690

Data Index: 509352 (C4 (Web Text))

Max Activating Token Index: 26

Text #1

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.8878. Min Act: -0.1696

Data Index: 379728 (C4 (Web Text))

Max Activating Token Index: 23

Text #2

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.8252. Min Act: -0.1700

Data Index: 948402 (C4 (Web Text))

Max Activating Token Index: 57

Text #3

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7627. Min Act: -0.1700

Data Index: 942159 (C4 (Web Text))

Max Activating Token Index: 306

Text #4

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7627. Min Act: -0.1700

Data Index: 730077 (C4 (Web Text))

Max Activating Token Index: 925

Text #5

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7627. Min Act: -0.1700

Data Index: 514072 (C4 (Web Text))

Max Activating Token Index: 681

Text #6

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7940. Min Act: -0.1698

Data Index: 281554 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7315. Min Act: -0.1700

Data Index: 65104 (C4 (Web Text))

Max Activating Token Index: 605

Text #8

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7315. Min Act: -0.1700

Data Index: 127926 (C4 (Web Text))

Max Activating Token Index: 107

Text #9

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7315. Min Act: -0.1700

Data Index: 1121111 (C4 (Web Text))

Max Activating Token Index: 98

Text #10

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7315. Min Act: -0.1694

Data Index: 570758 (C4 (Web Text))

Max Activating Token Index: 1001

Text #11

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.6690. Min Act: -0.1699

Data Index: 909540 (C4 (Web Text))

Max Activating Token Index: 630

Text #12

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7002. Min Act: -0.1700

Data Index: 721305 (C4 (Web Text))

Max Activating Token Index: 702

Text #13

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7002. Min Act: -0.1697

Data Index: 813447 (C4 (Web Text))

Max Activating Token Index: 241

Text #14

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7002. Min Act: -0.1699

Data Index: 350014 (C4 (Web Text))

Max Activating Token Index: 692

Text #15

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.7002. Min Act: -0.1700

Data Index: 828925 (C4 (Web Text))

Max Activating Token Index: 601

Text #16

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.6690. Min Act: -0.1700

Data Index: 397371 (C4 (Web Text))

Max Activating Token Index: 641

Text #17

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.6690. Min Act: -0.1700

Data Index: 1027728 (C4 (Web Text))

Max Activating Token Index: 107

Text #18

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.6377. Min Act: -0.1700

Data Index: 858294 (C4 (Web Text))

Max Activating Token Index: 820

Text #19

Max Range: 5.1065. Min Range: -5.1065

Max Act: 4.6377. Min Act: -0.1700

Data Index: 693607 (C4 (Web Text))

Max Activating Token Index: 811