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

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

Neuron 1271 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.8897. Min Act: -0.1700

Data Index: 479232 (C4 (Web Text))

Max Activating Token Index: 534

Text #1

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.8897. Min Act: -0.1700

Data Index: 83804 (C4 (Web Text))

Max Activating Token Index: 136

Text #2

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.8271. Min Act: -0.1700

Data Index: 319562 (C4 (Web Text))

Max Activating Token Index: 564

Text #3

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 435322 (C4 (Web Text))

Max Activating Token Index: 24

Text #4

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 435814 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 645019 (C4 (Web Text))

Max Activating Token Index: 848

Text #6

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 658914 (C4 (Web Text))

Max Activating Token Index: 604

Text #7

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 1647201 (Python Code)

Max Activating Token Index: 74

Text #8

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 1574880 (Python Code)

Max Activating Token Index: 916

Text #9

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 13627 (C4 (Web Text))

Max Activating Token Index: 169

Text #10

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 721271 (C4 (Web Text))

Max Activating Token Index: 159

Text #11

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 580950 (C4 (Web Text))

Max Activating Token Index: 947

Text #12

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7959. Min Act: -0.1700

Data Index: 782190 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 915800 (C4 (Web Text))

Max Activating Token Index: 41

Text #14

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7333. Min Act: -0.1700

Data Index: 1067625 (C4 (Web Text))

Max Activating Token Index: 432

Text #15

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7020. Min Act: -0.1700

Data Index: 421391 (C4 (Web Text))

Max Activating Token Index: 614

Text #16

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7333. Min Act: -0.1700

Data Index: 371816 (C4 (Web Text))

Max Activating Token Index: 941

Text #17

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7020. Min Act: -0.1700

Data Index: 10646 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7646. Min Act: -0.1700

Data Index: 1503951 (Python Code)

Max Activating Token Index: 768

Text #19

Max Range: 3.8897. Min Range: -3.8897

Max Act: 3.7020. Min Act: -0.1700

Data Index: 495615 (C4 (Web Text))

Max Activating Token Index: 955