Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4196 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.8824. Min Range: -2.8824

Max Act: 2.8824. Min Act: -0.0000

Data Index: 119788 (C4 (Web Text))

Max Activating Token Index: 288

Text #1

Max Range: 2.8824. Min Range: -2.8824

Max Act: 2.3421. Min Act: -0.0000

Data Index: 1142037 (C4 (Web Text))

Max Activating Token Index: 1016

Text #2

Max Range: 2.8824. Min Range: -2.8824

Max Act: 2.2766. Min Act: -0.0000

Data Index: 622357 (C4 (Web Text))

Max Activating Token Index: 15

Text #3

Max Range: 2.8824. Min Range: -2.8824

Max Act: 2.0783. Min Act: -0.0000

Data Index: 899777 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.9557. Min Act: -0.0000

Data Index: 669031 (C4 (Web Text))

Max Activating Token Index: 902

Text #5

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.8261. Min Act: -0.0000

Data Index: 794377 (C4 (Web Text))

Max Activating Token Index: 240

Text #6

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.7486. Min Act: -0.0000

Data Index: 1577516 (Python Code)

Max Activating Token Index: 343

Text #7

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.8637. Min Act: -0.0000

Data Index: 1433869 (Python Code)

Max Activating Token Index: 374

Text #8

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.6382. Min Act: -0.0000

Data Index: 231337 (C4 (Web Text))

Max Activating Token Index: 701

Text #9

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.5872. Min Act: -0.0000

Data Index: 1522457 (Python Code)

Max Activating Token Index: 352

Text #10

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.5871. Min Act: -0.0000

Data Index: 1555115 (Python Code)

Max Activating Token Index: 104

Text #11

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.4833. Min Act: -0.0000

Data Index: 1594683 (Python Code)

Max Activating Token Index: 49

Text #12

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.3940. Min Act: -0.0000

Data Index: 1608516 (Python Code)

Max Activating Token Index: 840

Text #13

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.4426. Min Act: -0.0000

Data Index: 870623 (C4 (Web Text))

Max Activating Token Index: 634

Text #14

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.4130. Min Act: -0.0000

Data Index: 1562228 (Python Code)

Max Activating Token Index: 275

Text #15

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.3130. Min Act: -0.0000

Data Index: 1379349 (Python Code)

Max Activating Token Index: 478

Text #16

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.2870. Min Act: -0.0000

Data Index: 1551259 (Python Code)

Max Activating Token Index: 191

Text #17

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.2969. Min Act: -0.0000

Data Index: 1617837 (Python Code)

Max Activating Token Index: 262

Text #18

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.2842. Min Act: -0.0000

Data Index: 1464737 (Python Code)

Max Activating Token Index: 642

Text #19

Max Range: 2.8824. Min Range: -2.8824

Max Act: 1.3657. Min Act: -0.0000

Data Index: 1441475 (Python Code)

Max Activating Token Index: 506