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

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

Neuron 90 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: 5.2303. Min Range: -5.2303

Max Act: 5.2303. Min Act: -0.1700

Data Index: 206575 (C4 (Web Text))

Max Activating Token Index: 170

Text #1

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1990. Min Act: -0.1700

Data Index: 328782 (C4 (Web Text))

Max Activating Token Index: 332

Text #2

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1365. Min Act: -0.1700

Data Index: 293911 (C4 (Web Text))

Max Activating Token Index: 141

Text #3

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0740. Min Act: -0.1700

Data Index: 46842 (C4 (Web Text))

Max Activating Token Index: 418

Text #4

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1053. Min Act: -0.1700

Data Index: 937789 (C4 (Web Text))

Max Activating Token Index: 32

Text #5

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1365. Min Act: -0.1700

Data Index: 1205050 (C4 (Web Text))

Max Activating Token Index: 203

Text #6

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1053. Min Act: -0.1700

Data Index: 581374 (C4 (Web Text))

Max Activating Token Index: 554

Text #7

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1053. Min Act: -0.1700

Data Index: 773562 (C4 (Web Text))

Max Activating Token Index: 45

Text #8

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0428. Min Act: -0.1700

Data Index: 30054 (C4 (Web Text))

Max Activating Token Index: 259

Text #9

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.1053. Min Act: -0.1700

Data Index: 1035828 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0115. Min Act: -0.1700

Data Index: 874319 (C4 (Web Text))

Max Activating Token Index: 506

Text #11

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0740. Min Act: -0.1700

Data Index: 1028172 (C4 (Web Text))

Max Activating Token Index: 687

Text #12

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0428. Min Act: -0.1700

Data Index: 783834 (C4 (Web Text))

Max Activating Token Index: 798

Text #13

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0740. Min Act: -0.1700

Data Index: 810920 (C4 (Web Text))

Max Activating Token Index: 757

Text #14

Max Range: 5.2303. Min Range: -5.2303

Max Act: 4.9803. Min Act: -0.1700

Data Index: 285942 (C4 (Web Text))

Max Activating Token Index: 775

Text #15

Max Range: 5.2303. Min Range: -5.2303

Max Act: 4.9803. Min Act: -0.1700

Data Index: 1161824 (C4 (Web Text))

Max Activating Token Index: 829

Text #16

Max Range: 5.2303. Min Range: -5.2303

Max Act: 4.9803. Min Act: -0.1700

Data Index: 259640 (C4 (Web Text))

Max Activating Token Index: 716

Text #17

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0115. Min Act: -0.1700

Data Index: 641311 (C4 (Web Text))

Max Activating Token Index: 305

Text #18

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0115. Min Act: -0.1700

Data Index: 630289 (C4 (Web Text))

Max Activating Token Index: 561

Text #19

Max Range: 5.2303. Min Range: -5.2303

Max Act: 5.0428. Min Act: -0.1700

Data Index: 831871 (C4 (Web Text))

Max Activating Token Index: 51