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

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

Neuron 4685 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.4740. Min Act: -0.0003

Data Index: 662911 (C4 (Web Text))

Max Activating Token Index: 810

Text #1

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.4606. Min Act: -0.0002

Data Index: 799426 (C4 (Web Text))

Max Activating Token Index: 235

Text #2

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.4450. Min Act: -0.0003

Data Index: 628084 (C4 (Web Text))

Max Activating Token Index: 65

Text #3

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.4320. Min Act: -0.0003

Data Index: 728502 (C4 (Web Text))

Max Activating Token Index: 385

Text #4

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3857. Min Act: -0.0003

Data Index: 271075 (C4 (Web Text))

Max Activating Token Index: 683

Text #5

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3927. Min Act: -0.0002

Data Index: 448426 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3667. Min Act: -0.0002

Data Index: 222910 (C4 (Web Text))

Max Activating Token Index: 665

Text #7

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3736. Min Act: -0.0003

Data Index: 33796 (C4 (Web Text))

Max Activating Token Index: 656

Text #8

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3722. Min Act: -0.0003

Data Index: 36834 (C4 (Web Text))

Max Activating Token Index: 172

Text #9

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3590. Min Act: -0.0003

Data Index: 1181104 (C4 (Web Text))

Max Activating Token Index: 342

Text #10

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3165. Min Act: -0.0003

Data Index: 1356168 (C4 (Web Text))

Max Activating Token Index: 738

Text #11

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3135. Min Act: -0.0003

Data Index: 958358 (C4 (Web Text))

Max Activating Token Index: 243

Text #12

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3207. Min Act: -0.0003

Data Index: 772818 (C4 (Web Text))

Max Activating Token Index: 565

Text #13

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3043. Min Act: -0.0003

Data Index: 55774 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3111. Min Act: -0.0003

Data Index: 213703 (C4 (Web Text))

Max Activating Token Index: 950

Text #15

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3078. Min Act: -0.0002

Data Index: 975671 (C4 (Web Text))

Max Activating Token Index: 575

Text #16

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.2959. Min Act: -0.0003

Data Index: 396645 (C4 (Web Text))

Max Activating Token Index: 321

Text #17

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3162. Min Act: -0.0003

Data Index: 418901 (C4 (Web Text))

Max Activating Token Index: 429

Text #18

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3060. Min Act: -0.0003

Data Index: 70737 (C4 (Web Text))

Max Activating Token Index: 502

Text #19

Max Range: 0.4740. Min Range: -0.4740

Max Act: 0.3050. Min Act: -0.0002

Data Index: 1064875 (C4 (Web Text))

Max Activating Token Index: 622