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

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

Neuron 3177 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.6093. Min Range: -4.6093

Max Act: 4.5399. Min Act: -0.0000

Data Index: 290719 (C4 (Web Text))

Max Activating Token Index: 1005

Text #1

Max Range: 4.6093. Min Range: -4.6093

Max Act: 4.6093. Min Act: -0.0000

Data Index: 215407 (C4 (Web Text))

Max Activating Token Index: 221

Text #2

Max Range: 4.6093. Min Range: -4.6093

Max Act: 4.4523. Min Act: -0.0000

Data Index: 1201108 (C4 (Web Text))

Max Activating Token Index: 455

Text #3

Max Range: 4.6093. Min Range: -4.6093

Max Act: 4.5921. Min Act: -0.0000

Data Index: 683615 (C4 (Web Text))

Max Activating Token Index: 916

Text #4

Max Range: 4.6093. Min Range: -4.6093

Max Act: 4.5150. Min Act: -0.0000

Data Index: 787044 (C4 (Web Text))

Max Activating Token Index: 670

Text #5

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.8795. Min Act: -0.0000

Data Index: 168867 (C4 (Web Text))

Max Activating Token Index: 803

Text #6

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.7271. Min Act: -0.0000

Data Index: 277853 (C4 (Web Text))

Max Activating Token Index: 822

Text #7

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.7419. Min Act: -0.0000

Data Index: 335605 (C4 (Web Text))

Max Activating Token Index: 722

Text #8

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.5350. Min Act: -0.0000

Data Index: 724160 (C4 (Web Text))

Max Activating Token Index: 724

Text #9

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.3019. Min Act: -0.0000

Data Index: 1420469 (Python Code)

Max Activating Token Index: 938

Text #10

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.3666. Min Act: -0.0000

Data Index: 647484 (C4 (Web Text))

Max Activating Token Index: 341

Text #11

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.1323. Min Act: -0.0000

Data Index: 146897 (C4 (Web Text))

Max Activating Token Index: 614

Text #12

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.0820. Min Act: -0.0000

Data Index: 835149 (C4 (Web Text))

Max Activating Token Index: 381

Text #13

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.0031. Min Act: -0.0000

Data Index: 510614 (C4 (Web Text))

Max Activating Token Index: 385

Text #14

Max Range: 4.6093. Min Range: -4.6093

Max Act: 2.8222. Min Act: -0.0000

Data Index: 675063 (C4 (Web Text))

Max Activating Token Index: 1009

Text #15

Max Range: 4.6093. Min Range: -4.6093

Max Act: 2.7998. Min Act: -0.0000

Data Index: 1650044 (Python Code)

Max Activating Token Index: 1008

Text #16

Max Range: 4.6093. Min Range: -4.6093

Max Act: 2.9508. Min Act: -0.0000

Data Index: 311894 (C4 (Web Text))

Max Activating Token Index: 482

Text #17

Max Range: 4.6093. Min Range: -4.6093

Max Act: 2.7939. Min Act: -0.0000

Data Index: 375623 (C4 (Web Text))

Max Activating Token Index: 820

Text #18

Max Range: 4.6093. Min Range: -4.6093

Max Act: 2.9209. Min Act: -0.0000

Data Index: 1263774 (C4 (Web Text))

Max Activating Token Index: 331

Text #19

Max Range: 4.6093. Min Range: -4.6093

Max Act: 3.0409. Min Act: -0.0000

Data Index: 658601 (C4 (Web Text))

Max Activating Token Index: 788