Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1577 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.9535. Min Act: -0.1700

Data Index: 1266385 (C4 (Web Text))

Max Activating Token Index: 410

Text #1

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.9535. Min Act: -0.1700

Data Index: 1302571 (C4 (Web Text))

Max Activating Token Index: 959

Text #2

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.9223. Min Act: -0.1699

Data Index: 698601 (C4 (Web Text))

Max Activating Token Index: 124

Text #3

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.9223. Min Act: -0.1700

Data Index: 1265645 (C4 (Web Text))

Max Activating Token Index: 58

Text #4

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.8910. Min Act: -0.1700

Data Index: 1542572 (Python Code)

Max Activating Token Index: 992

Text #5

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.8910. Min Act: -0.1700

Data Index: 198299 (C4 (Web Text))

Max Activating Token Index: 632

Text #6

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.8128. Min Act: -0.1700

Data Index: 1050091 (C4 (Web Text))

Max Activating Token Index: 19

Text #7

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.8284. Min Act: -0.1700

Data Index: 1106288 (C4 (Web Text))

Max Activating Token Index: 1018

Text #8

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.8441. Min Act: -0.1700

Data Index: 1555069 (Python Code)

Max Activating Token Index: 306

Text #9

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7659. Min Act: -0.1700

Data Index: 1291180 (C4 (Web Text))

Max Activating Token Index: 330

Text #10

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7815. Min Act: -0.1700

Data Index: 265551 (C4 (Web Text))

Max Activating Token Index: 456

Text #11

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7659. Min Act: -0.1700

Data Index: 332958 (C4 (Web Text))

Max Activating Token Index: 963

Text #12

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7502. Min Act: -0.1700

Data Index: 512094 (C4 (Web Text))

Max Activating Token Index: 782

Text #13

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7346. Min Act: -0.1699

Data Index: 714831 (C4 (Web Text))

Max Activating Token Index: 754

Text #14

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7502. Min Act: -0.1700

Data Index: 78034 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7346. Min Act: -0.1700

Data Index: 241866 (C4 (Web Text))

Max Activating Token Index: 179

Text #16

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7033. Min Act: -0.1700

Data Index: 118588 (C4 (Web Text))

Max Activating Token Index: 908

Text #17

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7189. Min Act: -0.1700

Data Index: 1247541 (C4 (Web Text))

Max Activating Token Index: 384

Text #18

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7189. Min Act: -0.1700

Data Index: 930872 (C4 (Web Text))

Max Activating Token Index: 599

Text #19

Max Range: 3.9535. Min Range: -3.9535

Max Act: 3.7033. Min Act: -0.1700

Data Index: 1292675 (C4 (Web Text))

Max Activating Token Index: 878