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

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

Neuron 641 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3209. Min Range: -5.3209

Max Act: 5.3209. Min Act: -0.1700

Data Index: 445365 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 5.3209. Min Range: -5.3209

Max Act: 5.1647. Min Act: -0.1700

Data Index: 490549 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 5.3209. Min Range: -5.3209

Max Act: 5.1022. Min Act: -0.1700

Data Index: 768453 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.7272. Min Act: -0.1700

Data Index: 81730 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.6959. Min Act: -0.1700

Data Index: 651457 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.6021. Min Act: -0.1700

Data Index: 686284 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4146. Min Act: -0.1700

Data Index: 808669 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 459298 (C4 (Web Text))

Max Activating Token Index: 652

Text #8

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 339817 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 1328759 (C4 (Web Text))

Max Activating Token Index: 9

Text #10

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 795414 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4146. Min Act: -0.1700

Data Index: 125098 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.3521. Min Act: -0.1700

Data Index: 70162 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 651720 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4459. Min Act: -0.1700

Data Index: 508388 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.3834. Min Act: -0.1700

Data Index: 548160 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4146. Min Act: -0.1700

Data Index: 889866 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.4146. Min Act: -0.1699

Data Index: 720511 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.1959. Min Act: -0.1700

Data Index: 474938 (C4 (Web Text))

Max Activating Token Index: 435

Text #19

Max Range: 5.3209. Min Range: -5.3209

Max Act: 4.2584. Min Act: -0.1700

Data Index: 89420 (C4 (Web Text))

Max Activating Token Index: 1023