Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1024 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1669. Min Range: -5.1669

Max Act: 5.1669. Min Act: -0.1700

Data Index: 780565 (C4 (Web Text))

Max Activating Token Index: 984

Text #1

Max Range: 5.1669. Min Range: -5.1669

Max Act: 5.1044. Min Act: -0.1700

Data Index: 668857 (C4 (Web Text))

Max Activating Token Index: 426

Text #2

Max Range: 5.1669. Min Range: -5.1669

Max Act: 5.0732. Min Act: -0.1700

Data Index: 194009 (C4 (Web Text))

Max Activating Token Index: 154

Text #3

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9794. Min Act: -0.1700

Data Index: 1359839 (C4 (Web Text))

Max Activating Token Index: 1004

Text #4

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 882542 (C4 (Web Text))

Max Activating Token Index: 512

Text #5

Max Range: 5.1669. Min Range: -5.1669

Max Act: 5.0107. Min Act: -0.1700

Data Index: 1350309 (C4 (Web Text))

Max Activating Token Index: 705

Text #6

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9169. Min Act: -0.1700

Data Index: 26820 (C4 (Web Text))

Max Activating Token Index: 482

Text #7

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 707363 (C4 (Web Text))

Max Activating Token Index: 675

Text #8

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.8857. Min Act: -0.1700

Data Index: 1191593 (C4 (Web Text))

Max Activating Token Index: 596

Text #9

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 886623 (C4 (Web Text))

Max Activating Token Index: 831

Text #10

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9794. Min Act: -0.1700

Data Index: 599524 (C4 (Web Text))

Max Activating Token Index: 655

Text #11

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.8544. Min Act: -0.1700

Data Index: 535884 (C4 (Web Text))

Max Activating Token Index: 771

Text #12

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 629085 (C4 (Web Text))

Max Activating Token Index: 310

Text #13

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9169. Min Act: -0.1700

Data Index: 595287 (C4 (Web Text))

Max Activating Token Index: 982

Text #14

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 428435 (C4 (Web Text))

Max Activating Token Index: 288

Text #15

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9482. Min Act: -0.1700

Data Index: 1179299 (C4 (Web Text))

Max Activating Token Index: 986

Text #16

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.7919. Min Act: -0.1700

Data Index: 800091 (C4 (Web Text))

Max Activating Token Index: 876

Text #17

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.8857. Min Act: -0.1700

Data Index: 748678 (C4 (Web Text))

Max Activating Token Index: 851

Text #18

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.8857. Min Act: -0.1700

Data Index: 970420 (C4 (Web Text))

Max Activating Token Index: 433

Text #19

Max Range: 5.1669. Min Range: -5.1669

Max Act: 4.9169. Min Act: -0.1700

Data Index: 42994 (C4 (Web Text))

Max Activating Token Index: 824