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

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

Neuron 2999 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4097. Min Range: -2.4097

Max Act: 2.4097. Min Act: -0.0000

Data Index: 342329 (C4 (Web Text))

Max Activating Token Index: 270

Text #1

Max Range: 2.4097. Min Range: -2.4097

Max Act: 2.1788. Min Act: -0.0000

Data Index: 871695 (C4 (Web Text))

Max Activating Token Index: 506

Text #2

Max Range: 2.4097. Min Range: -2.4097

Max Act: 2.0353. Min Act: -0.0000

Data Index: 949555 (C4 (Web Text))

Max Activating Token Index: 154

Text #3

Max Range: 2.4097. Min Range: -2.4097

Max Act: 2.0069. Min Act: -0.0000

Data Index: 1599254 (Python Code)

Max Activating Token Index: 337

Text #4

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.9211. Min Act: -0.0000

Data Index: 858113 (C4 (Web Text))

Max Activating Token Index: 886

Text #5

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.9232. Min Act: -0.0000

Data Index: 1046797 (C4 (Web Text))

Max Activating Token Index: 491

Text #6

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.9129. Min Act: -0.0000

Data Index: 1486809 (Python Code)

Max Activating Token Index: 692

Text #7

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.9473. Min Act: -0.0000

Data Index: 1501733 (Python Code)

Max Activating Token Index: 163

Text #8

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8904. Min Act: -0.0000

Data Index: 814600 (C4 (Web Text))

Max Activating Token Index: 568

Text #9

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8181. Min Act: -0.0000

Data Index: 1581300 (Python Code)

Max Activating Token Index: 67

Text #10

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7942. Min Act: -0.0000

Data Index: 1448992 (Python Code)

Max Activating Token Index: 798

Text #11

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8448. Min Act: -0.0000

Data Index: 215206 (C4 (Web Text))

Max Activating Token Index: 297

Text #12

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8316. Min Act: -0.0000

Data Index: 281361 (C4 (Web Text))

Max Activating Token Index: 881

Text #13

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7610. Min Act: -0.0000

Data Index: 1399076 (Python Code)

Max Activating Token Index: 944

Text #14

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8011. Min Act: -0.0000

Data Index: 1027876 (C4 (Web Text))

Max Activating Token Index: 380

Text #15

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7723. Min Act: -0.0000

Data Index: 864901 (C4 (Web Text))

Max Activating Token Index: 285

Text #16

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.8251. Min Act: -0.0000

Data Index: 374678 (C4 (Web Text))

Max Activating Token Index: 497

Text #17

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7512. Min Act: -0.0000

Data Index: 286823 (C4 (Web Text))

Max Activating Token Index: 39

Text #18

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7412. Min Act: -0.0000

Data Index: 204663 (C4 (Web Text))

Max Activating Token Index: 792

Text #19

Max Range: 2.4097. Min Range: -2.4097

Max Act: 1.7312. Min Act: -0.0000

Data Index: 1491963 (Python Code)

Max Activating Token Index: 249