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

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

Neuron 2500 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.5390. Min Act: -0.0002

Data Index: 58962 (C4 (Web Text))

Max Activating Token Index: 2

Text #1

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.5185. Min Act: -0.0002

Data Index: 816714 (C4 (Web Text))

Max Activating Token Index: 127

Text #2

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4731. Min Act: -0.0002

Data Index: 551198 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4487. Min Act: -0.0002

Data Index: 349026 (C4 (Web Text))

Max Activating Token Index: 755

Text #4

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4346. Min Act: -0.0002

Data Index: 147128 (C4 (Web Text))

Max Activating Token Index: 7

Text #5

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4131. Min Act: -0.0002

Data Index: 361054 (C4 (Web Text))

Max Activating Token Index: 3

Text #6

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4578. Min Act: -0.0002

Data Index: 35094 (C4 (Web Text))

Max Activating Token Index: 6

Text #7

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4375. Min Act: -0.0002

Data Index: 897655 (C4 (Web Text))

Max Activating Token Index: 78

Text #8

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4181. Min Act: -0.0002

Data Index: 169797 (C4 (Web Text))

Max Activating Token Index: 2

Text #9

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4635. Min Act: -0.0002

Data Index: 851313 (C4 (Web Text))

Max Activating Token Index: 39

Text #10

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3939. Min Act: -0.0002

Data Index: 1302501 (C4 (Web Text))

Max Activating Token Index: 260

Text #11

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4021. Min Act: -0.0002

Data Index: 213989 (C4 (Web Text))

Max Activating Token Index: 3

Text #12

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.4107. Min Act: -0.0002

Data Index: 652596 (C4 (Web Text))

Max Activating Token Index: 567

Text #13

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3904. Min Act: -0.0002

Data Index: 52500 (C4 (Web Text))

Max Activating Token Index: 372

Text #14

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3587. Min Act: -0.0002

Data Index: 439669 (C4 (Web Text))

Max Activating Token Index: 703

Text #15

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3413. Min Act: -0.0002

Data Index: 1121770 (C4 (Web Text))

Max Activating Token Index: 447

Text #16

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3488. Min Act: -0.0002

Data Index: 1177243 (C4 (Web Text))

Max Activating Token Index: 41

Text #17

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3673. Min Act: -0.0002

Data Index: 295210 (C4 (Web Text))

Max Activating Token Index: 203

Text #18

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.2991. Min Act: -0.0002

Data Index: 194818 (C4 (Web Text))

Max Activating Token Index: 297

Text #19

Max Range: 2.5390. Min Range: -2.5390

Max Act: 2.3008. Min Act: -0.0002

Data Index: 1067023 (C4 (Web Text))

Max Activating Token Index: 767