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

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

Neuron 4686 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.8858. Min Act: -0.0002

Data Index: 243421 (C4 (Web Text))

Max Activating Token Index: 53

Text #1

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.8468. Min Act: -0.0002

Data Index: 611910 (C4 (Web Text))

Max Activating Token Index: 529

Text #2

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.8160. Min Act: -0.0002

Data Index: 1162485 (C4 (Web Text))

Max Activating Token Index: 120

Text #3

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.8280. Min Act: -0.0002

Data Index: 1289166 (C4 (Web Text))

Max Activating Token Index: 873

Text #4

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7866. Min Act: -0.0002

Data Index: 698346 (C4 (Web Text))

Max Activating Token Index: 75

Text #5

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7398. Min Act: -0.0002

Data Index: 583578 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7166. Min Act: -0.0002

Data Index: 394062 (C4 (Web Text))

Max Activating Token Index: 358

Text #7

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7886. Min Act: -0.0002

Data Index: 363097 (C4 (Web Text))

Max Activating Token Index: 813

Text #8

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7348. Min Act: -0.0002

Data Index: 166585 (C4 (Web Text))

Max Activating Token Index: 953

Text #9

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7755. Min Act: -0.0002

Data Index: 998438 (C4 (Web Text))

Max Activating Token Index: 841

Text #10

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7705. Min Act: -0.0002

Data Index: 20600 (C4 (Web Text))

Max Activating Token Index: 370

Text #11

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.8116. Min Act: -0.0002

Data Index: 340107 (C4 (Web Text))

Max Activating Token Index: 340

Text #12

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7894. Min Act: -0.0002

Data Index: 916679 (C4 (Web Text))

Max Activating Token Index: 173

Text #13

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7061. Min Act: -0.0002

Data Index: 952977 (C4 (Web Text))

Max Activating Token Index: 883

Text #14

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7027. Min Act: -0.0002

Data Index: 932611 (C4 (Web Text))

Max Activating Token Index: 350

Text #15

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7063. Min Act: -0.0002

Data Index: 300363 (C4 (Web Text))

Max Activating Token Index: 856

Text #16

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7231. Min Act: -0.0002

Data Index: 1047309 (C4 (Web Text))

Max Activating Token Index: 393

Text #17

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7638. Min Act: -0.0002

Data Index: 365067 (C4 (Web Text))

Max Activating Token Index: 21

Text #18

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7543. Min Act: -0.0002

Data Index: 15667 (C4 (Web Text))

Max Activating Token Index: 547

Text #19

Max Range: 0.8858. Min Range: -0.8858

Max Act: 0.7554. Min Act: -0.0002

Data Index: 41421 (C4 (Web Text))

Max Activating Token Index: 940