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

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

Neuron 3040 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.2803. Min Range: -2.2803

Max Act: 2.1652. Min Act: -0.0000

Data Index: 1537393 (Python Code)

Max Activating Token Index: 742

Text #1

Max Range: 2.2803. Min Range: -2.2803

Max Act: 2.2803. Min Act: -0.0000

Data Index: 1406464 (Python Code)

Max Activating Token Index: 496

Text #2

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.9681. Min Act: -0.0000

Data Index: 942383 (C4 (Web Text))

Max Activating Token Index: 587

Text #3

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.9576. Min Act: -0.0000

Data Index: 1252421 (C4 (Web Text))

Max Activating Token Index: 648

Text #4

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.9556. Min Act: -0.0000

Data Index: 1498836 (Python Code)

Max Activating Token Index: 563

Text #5

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.9499. Min Act: -0.0000

Data Index: 585933 (C4 (Web Text))

Max Activating Token Index: 28

Text #6

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.8253. Min Act: -0.0000

Data Index: 777248 (C4 (Web Text))

Max Activating Token Index: 132

Text #7

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.9201. Min Act: -0.0000

Data Index: 1570444 (Python Code)

Max Activating Token Index: 725

Text #8

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.8905. Min Act: -0.0000

Data Index: 246036 (C4 (Web Text))

Max Activating Token Index: 968

Text #9

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.8884. Min Act: -0.0000

Data Index: 1360314 (Python Code)

Max Activating Token Index: 860

Text #10

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.7139. Min Act: -0.0000

Data Index: 762239 (C4 (Web Text))

Max Activating Token Index: 428

Text #11

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.8016. Min Act: -0.0000

Data Index: 1135595 (C4 (Web Text))

Max Activating Token Index: 114

Text #12

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.7081. Min Act: -0.0000

Data Index: 1405736 (Python Code)

Max Activating Token Index: 445

Text #13

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6861. Min Act: -0.0000

Data Index: 1601080 (Python Code)

Max Activating Token Index: 357

Text #14

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6160. Min Act: -0.0000

Data Index: 1190051 (C4 (Web Text))

Max Activating Token Index: 626

Text #15

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6250. Min Act: -0.0000

Data Index: 1459655 (Python Code)

Max Activating Token Index: 276

Text #16

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6957. Min Act: -0.0000

Data Index: 223816 (C4 (Web Text))

Max Activating Token Index: 519

Text #17

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6900. Min Act: -0.0000

Data Index: 1530938 (Python Code)

Max Activating Token Index: 920

Text #18

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.6856. Min Act: -0.0000

Data Index: 343505 (C4 (Web Text))

Max Activating Token Index: 682

Text #19

Max Range: 2.2803. Min Range: -2.2803

Max Act: 1.5641. Min Act: -0.0000

Data Index: 206485 (C4 (Web Text))

Max Activating Token Index: 96