Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1486 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.5029. Min Act: -0.0001

Data Index: 480545 (C4 (Web Text))

Max Activating Token Index: 532

Text #1

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.5363. Min Act: -0.0001

Data Index: 619244 (C4 (Web Text))

Max Activating Token Index: 332

Text #2

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.4829. Min Act: -0.0001

Data Index: 531878 (C4 (Web Text))

Max Activating Token Index: 939

Text #3

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3172. Min Act: -0.0001

Data Index: 35223 (C4 (Web Text))

Max Activating Token Index: 383

Text #4

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3532. Min Act: -0.0001

Data Index: 1024165 (C4 (Web Text))

Max Activating Token Index: 815

Text #5

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2603. Min Act: -0.0001

Data Index: 358329 (C4 (Web Text))

Max Activating Token Index: 985

Text #6

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3169. Min Act: -0.0001

Data Index: 1256380 (C4 (Web Text))

Max Activating Token Index: 14

Text #7

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3545. Min Act: -0.0001

Data Index: 1100882 (C4 (Web Text))

Max Activating Token Index: 37

Text #8

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3860. Min Act: -0.0001

Data Index: 974900 (C4 (Web Text))

Max Activating Token Index: 963

Text #9

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.3223. Min Act: -0.0001

Data Index: 1073027 (C4 (Web Text))

Max Activating Token Index: 801

Text #10

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2994. Min Act: -0.0001

Data Index: 853026 (C4 (Web Text))

Max Activating Token Index: 457

Text #11

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.1882. Min Act: -0.0001

Data Index: 153174 (C4 (Web Text))

Max Activating Token Index: 1002

Text #12

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2951. Min Act: -0.0001

Data Index: 79529 (C4 (Web Text))

Max Activating Token Index: 593

Text #13

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2134. Min Act: -0.0001

Data Index: 825557 (C4 (Web Text))

Max Activating Token Index: 663

Text #14

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.1724. Min Act: -0.0001

Data Index: 347230 (C4 (Web Text))

Max Activating Token Index: 980

Text #15

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2401. Min Act: -0.0001

Data Index: 648064 (C4 (Web Text))

Max Activating Token Index: 528

Text #16

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2124. Min Act: -0.0001

Data Index: 1450569 (Python Code)

Max Activating Token Index: 396

Text #17

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2003. Min Act: -0.0001

Data Index: 277179 (C4 (Web Text))

Max Activating Token Index: 738

Text #18

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.1909. Min Act: -0.0001

Data Index: 664339 (C4 (Web Text))

Max Activating Token Index: 837

Text #19

Max Range: 1.5363. Min Range: -1.5363

Max Act: 1.2261. Min Act: -0.0001

Data Index: 1109349 (C4 (Web Text))

Max Activating Token Index: 803