Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3501 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.6775. Min Act: -0.0000

Data Index: 1238757 (C4 (Web Text))

Max Activating Token Index: 694

Text #1

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5673. Min Act: -0.0000

Data Index: 494344 (C4 (Web Text))

Max Activating Token Index: 590

Text #2

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.6361. Min Act: 0.0000

Data Index: 1007574 (C4 (Web Text))

Max Activating Token Index: 203

Text #3

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.6049. Min Act: -0.0000

Data Index: 86329 (C4 (Web Text))

Max Activating Token Index: 667

Text #4

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5535. Min Act: -0.0000

Data Index: 1298299 (C4 (Web Text))

Max Activating Token Index: 283

Text #5

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5589. Min Act: -0.0000

Data Index: 636505 (C4 (Web Text))

Max Activating Token Index: 933

Text #6

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5735. Min Act: -0.0000

Data Index: 833643 (C4 (Web Text))

Max Activating Token Index: 474

Text #7

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5670. Min Act: -0.0000

Data Index: 1511491 (Python Code)

Max Activating Token Index: 299

Text #8

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5674. Min Act: 0.0001

Data Index: 52498 (C4 (Web Text))

Max Activating Token Index: 48

Text #9

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5593. Min Act: -0.0000

Data Index: 613398 (C4 (Web Text))

Max Activating Token Index: 556

Text #10

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5125. Min Act: -0.0000

Data Index: 26427 (C4 (Web Text))

Max Activating Token Index: 146

Text #11

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5387. Min Act: -0.0000

Data Index: 1449969 (Python Code)

Max Activating Token Index: 266

Text #12

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5340. Min Act: -0.0000

Data Index: 1547151 (Python Code)

Max Activating Token Index: 995

Text #13

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5626. Min Act: -0.0000

Data Index: 421268 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5857. Min Act: 0.0002

Data Index: 945155 (C4 (Web Text))

Max Activating Token Index: 231

Text #15

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5438. Min Act: 0.0002

Data Index: 571898 (C4 (Web Text))

Max Activating Token Index: 189

Text #16

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.4272. Min Act: -0.0000

Data Index: 857256 (C4 (Web Text))

Max Activating Token Index: 192

Text #17

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5528. Min Act: -0.0000

Data Index: 1164290 (C4 (Web Text))

Max Activating Token Index: 542

Text #18

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5050. Min Act: 0.0002

Data Index: 538109 (C4 (Web Text))

Max Activating Token Index: 12

Text #19

Max Range: 0.6775. Min Range: -0.6775

Max Act: 0.5515. Min Act: 0.0002

Data Index: 455522 (C4 (Web Text))

Max Activating Token Index: 126