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

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

Neuron 803 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3795. Min Act: -0.0001

Data Index: 1181405 (C4 (Web Text))

Max Activating Token Index: 610

Text #1

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3928. Min Act: -0.0001

Data Index: 405257 (C4 (Web Text))

Max Activating Token Index: 866

Text #2

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3338. Min Act: -0.0001

Data Index: 349718 (C4 (Web Text))

Max Activating Token Index: 262

Text #3

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3589. Min Act: -0.0001

Data Index: 490287 (C4 (Web Text))

Max Activating Token Index: 374

Text #4

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3077. Min Act: -0.0001

Data Index: 1465845 (Python Code)

Max Activating Token Index: 434

Text #5

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3402. Min Act: -0.0001

Data Index: 1048432 (C4 (Web Text))

Max Activating Token Index: 937

Text #6

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2885. Min Act: -0.0001

Data Index: 1386278 (Python Code)

Max Activating Token Index: 294

Text #7

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2913. Min Act: -0.0001

Data Index: 1500150 (Python Code)

Max Activating Token Index: 907

Text #8

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3061. Min Act: -0.0001

Data Index: 1540260 (Python Code)

Max Activating Token Index: 410

Text #9

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3129. Min Act: -0.0001

Data Index: 864604 (C4 (Web Text))

Max Activating Token Index: 416

Text #10

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2885. Min Act: -0.0001

Data Index: 1454515 (Python Code)

Max Activating Token Index: 530

Text #11

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2691. Min Act: -0.0001

Data Index: 1374841 (Python Code)

Max Activating Token Index: 320

Text #12

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2501. Min Act: -0.0001

Data Index: 307934 (C4 (Web Text))

Max Activating Token Index: 158

Text #13

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.3075. Min Act: -0.0001

Data Index: 960183 (C4 (Web Text))

Max Activating Token Index: 225

Text #14

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2731. Min Act: -0.0001

Data Index: 1451624 (Python Code)

Max Activating Token Index: 384

Text #15

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2261. Min Act: -0.0001

Data Index: 1029353 (C4 (Web Text))

Max Activating Token Index: 768

Text #16

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2858. Min Act: -0.0001

Data Index: 1477765 (Python Code)

Max Activating Token Index: 102

Text #17

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2769. Min Act: -0.0001

Data Index: 1414872 (Python Code)

Max Activating Token Index: 76

Text #18

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2980. Min Act: -0.0001

Data Index: 1492569 (Python Code)

Max Activating Token Index: 285

Text #19

Max Range: 0.3928. Min Range: -0.3928

Max Act: 0.2627. Min Act: -0.0001

Data Index: 515836 (C4 (Web Text))

Max Activating Token Index: 76