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

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

Neuron 366 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: 0.2485. Min Range: -0.2485

Max Act: 0.2371. Min Act: -0.0001

Data Index: 234702 (C4 (Web Text))

Max Activating Token Index: 401

Text #1

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2324. Min Act: -0.0001

Data Index: 977249 (C4 (Web Text))

Max Activating Token Index: 419

Text #2

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2399. Min Act: -0.0001

Data Index: 485670 (C4 (Web Text))

Max Activating Token Index: 683

Text #3

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2282. Min Act: -0.0001

Data Index: 478596 (C4 (Web Text))

Max Activating Token Index: 637

Text #4

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2395. Min Act: -0.0001

Data Index: 467973 (C4 (Web Text))

Max Activating Token Index: 448

Text #5

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2485. Min Act: -0.0001

Data Index: 345978 (C4 (Web Text))

Max Activating Token Index: 411

Text #6

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2070. Min Act: -0.0001

Data Index: 173602 (C4 (Web Text))

Max Activating Token Index: 311

Text #7

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2078. Min Act: -0.0000

Data Index: 1380966 (Python Code)

Max Activating Token Index: 1022

Text #8

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2072. Min Act: -0.0001

Data Index: 223605 (C4 (Web Text))

Max Activating Token Index: 399

Text #9

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2478. Min Act: -0.0001

Data Index: 129616 (C4 (Web Text))

Max Activating Token Index: 376

Text #10

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2104. Min Act: -0.0001

Data Index: 479447 (C4 (Web Text))

Max Activating Token Index: 875

Text #11

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1982. Min Act: -0.0001

Data Index: 1410631 (Python Code)

Max Activating Token Index: 1014

Text #12

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2087. Min Act: -0.0001

Data Index: 670793 (C4 (Web Text))

Max Activating Token Index: 121

Text #13

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2090. Min Act: -0.0000

Data Index: 1459698 (Python Code)

Max Activating Token Index: 1022

Text #14

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1964. Min Act: -0.0001

Data Index: 318196 (C4 (Web Text))

Max Activating Token Index: 853

Text #15

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2081. Min Act: -0.0001

Data Index: 840164 (C4 (Web Text))

Max Activating Token Index: 899

Text #16

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1975. Min Act: -0.0001

Data Index: 1525343 (Python Code)

Max Activating Token Index: 945

Text #17

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2019. Min Act: -0.0001

Data Index: 1626306 (Python Code)

Max Activating Token Index: 1007

Text #18

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.2082. Min Act: -0.0001

Data Index: 649746 (C4 (Web Text))

Max Activating Token Index: 417

Text #19

Max Range: 0.2485. Min Range: -0.2485

Max Act: 0.1964. Min Act: -0.0001

Data Index: 159329 (C4 (Web Text))

Max Activating Token Index: 678