Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1512 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.7731. Min Act: -0.0005

Data Index: 787493 (C4 (Web Text))

Max Activating Token Index: 767

Text #1

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.8750. Min Act: -0.0005

Data Index: 912702 (C4 (Web Text))

Max Activating Token Index: 183

Text #2

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.5704. Min Act: -0.0005

Data Index: 270819 (C4 (Web Text))

Max Activating Token Index: 544

Text #3

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.5128. Min Act: -0.0005

Data Index: 675699 (C4 (Web Text))

Max Activating Token Index: 765

Text #4

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.4529. Min Act: -0.0005

Data Index: 305272 (C4 (Web Text))

Max Activating Token Index: 68

Text #5

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.4990. Min Act: -0.0005

Data Index: 172319 (C4 (Web Text))

Max Activating Token Index: 931

Text #6

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3928. Min Act: -0.0005

Data Index: 770049 (C4 (Web Text))

Max Activating Token Index: 1015

Text #7

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3757. Min Act: -0.0005

Data Index: 614391 (C4 (Web Text))

Max Activating Token Index: 169

Text #8

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3270. Min Act: -0.0005

Data Index: 486597 (C4 (Web Text))

Max Activating Token Index: 101

Text #9

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3691. Min Act: -0.0005

Data Index: 3631 (C4 (Web Text))

Max Activating Token Index: 414

Text #10

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2546. Min Act: -0.0005

Data Index: 1132521 (C4 (Web Text))

Max Activating Token Index: 284

Text #11

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3650. Min Act: -0.0005

Data Index: 1161575 (C4 (Web Text))

Max Activating Token Index: 867

Text #12

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2695. Min Act: -0.0005

Data Index: 1284016 (C4 (Web Text))

Max Activating Token Index: 304

Text #13

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2986. Min Act: -0.0005

Data Index: 741092 (C4 (Web Text))

Max Activating Token Index: 486

Text #14

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.1766. Min Act: -0.0005

Data Index: 1525309 (Python Code)

Max Activating Token Index: 122

Text #15

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2560. Min Act: -0.0005

Data Index: 886227 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2731. Min Act: -0.0005

Data Index: 629512 (C4 (Web Text))

Max Activating Token Index: 1009

Text #17

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2653. Min Act: -0.0005

Data Index: 405512 (C4 (Web Text))

Max Activating Token Index: 1010

Text #18

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.2642. Min Act: -0.0005

Data Index: 347154 (C4 (Web Text))

Max Activating Token Index: 654

Text #19

Max Range: 1.8750. Min Range: -1.8750

Max Act: 1.3806. Min Act: -0.0005

Data Index: 1191457 (C4 (Web Text))

Max Activating Token Index: 282