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

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

Neuron 1483 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: 5.1968. Min Range: -5.1968

Max Act: 5.1968. Min Act: -0.1700

Data Index: 602510 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 5.1968. Min Range: -5.1968

Max Act: 5.0406. Min Act: -0.1700

Data Index: 661918 (C4 (Web Text))

Max Activating Token Index: 846

Text #2

Max Range: 5.1968. Min Range: -5.1968

Max Act: 5.1031. Min Act: -0.1700

Data Index: 604009 (C4 (Web Text))

Max Activating Token Index: 1002

Text #3

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9468. Min Act: -0.1700

Data Index: 151612 (C4 (Web Text))

Max Activating Token Index: 907

Text #4

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9781. Min Act: -0.1700

Data Index: 295004 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 5.1968. Min Range: -5.1968

Max Act: 5.0093. Min Act: -0.1700

Data Index: 145317 (C4 (Web Text))

Max Activating Token Index: 671

Text #6

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9468. Min Act: -0.1700

Data Index: 1278043 (C4 (Web Text))

Max Activating Token Index: 591

Text #7

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9468. Min Act: -0.1700

Data Index: 242674 (C4 (Web Text))

Max Activating Token Index: 738

Text #8

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9468. Min Act: -0.1700

Data Index: 702307 (C4 (Web Text))

Max Activating Token Index: 312

Text #9

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8843. Min Act: -0.1700

Data Index: 21326 (C4 (Web Text))

Max Activating Token Index: 805

Text #10

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9468. Min Act: -0.1700

Data Index: 105490 (C4 (Web Text))

Max Activating Token Index: 884

Text #11

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.9156. Min Act: -0.1700

Data Index: 549687 (C4 (Web Text))

Max Activating Token Index: 809

Text #12

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8531. Min Act: -0.1700

Data Index: 1125490 (C4 (Web Text))

Max Activating Token Index: 990

Text #13

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8218. Min Act: -0.1700

Data Index: 1429413 (Python Code)

Max Activating Token Index: 765

Text #14

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8218. Min Act: -0.1700

Data Index: 1452336 (Python Code)

Max Activating Token Index: 420

Text #15

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8843. Min Act: -0.1700

Data Index: 879272 (C4 (Web Text))

Max Activating Token Index: 626

Text #16

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8843. Min Act: -0.1700

Data Index: 128824 (C4 (Web Text))

Max Activating Token Index: 793

Text #17

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8531. Min Act: -0.1700

Data Index: 274115 (C4 (Web Text))

Max Activating Token Index: 571

Text #18

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8531. Min Act: -0.1700

Data Index: 186417 (C4 (Web Text))

Max Activating Token Index: 634

Text #19

Max Range: 5.1968. Min Range: -5.1968

Max Act: 4.8218. Min Act: -0.1700

Data Index: 801974 (C4 (Web Text))

Max Activating Token Index: 384