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

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

Neuron 103 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.8542. Min Act: -0.0000

Data Index: 659525 (C4 (Web Text))

Max Activating Token Index: 615

Text #1

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.6690. Min Act: -0.0000

Data Index: 613238 (C4 (Web Text))

Max Activating Token Index: 429

Text #2

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5563. Min Act: -0.0000

Data Index: 761395 (C4 (Web Text))

Max Activating Token Index: 518

Text #3

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5674. Min Act: -0.0000

Data Index: 12318 (C4 (Web Text))

Max Activating Token Index: 195

Text #4

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5547. Min Act: -0.0000

Data Index: 106194 (C4 (Web Text))

Max Activating Token Index: 505

Text #5

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5422. Min Act: -0.0000

Data Index: 274383 (C4 (Web Text))

Max Activating Token Index: 1015

Text #6

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5168. Min Act: -0.0000

Data Index: 1121222 (C4 (Web Text))

Max Activating Token Index: 941

Text #7

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5106. Min Act: -0.0000

Data Index: 1622795 (Python Code)

Max Activating Token Index: 393

Text #8

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5096. Min Act: -0.0000

Data Index: 833353 (C4 (Web Text))

Max Activating Token Index: 320

Text #9

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4885. Min Act: -0.0000

Data Index: 1147100 (C4 (Web Text))

Max Activating Token Index: 181

Text #10

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4941. Min Act: -0.0000

Data Index: 972929 (C4 (Web Text))

Max Activating Token Index: 419

Text #11

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4960. Min Act: -0.0000

Data Index: 595099 (C4 (Web Text))

Max Activating Token Index: 52

Text #12

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4641. Min Act: -0.0000

Data Index: 1373103 (Python Code)

Max Activating Token Index: 10

Text #13

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4872. Min Act: -0.0000

Data Index: 1555388 (Python Code)

Max Activating Token Index: 685

Text #14

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4719. Min Act: -0.0000

Data Index: 1223834 (C4 (Web Text))

Max Activating Token Index: 766

Text #15

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4433. Min Act: -0.0000

Data Index: 659069 (C4 (Web Text))

Max Activating Token Index: 278

Text #16

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4877. Min Act: -0.0000

Data Index: 1256319 (C4 (Web Text))

Max Activating Token Index: 681

Text #17

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.5057. Min Act: -0.0000

Data Index: 332233 (C4 (Web Text))

Max Activating Token Index: 976

Text #18

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4865. Min Act: -0.0000

Data Index: 100601 (C4 (Web Text))

Max Activating Token Index: 950

Text #19

Max Range: 0.8542. Min Range: -0.8542

Max Act: 0.4669. Min Act: -0.0000

Data Index: 221381 (C4 (Web Text))

Max Activating Token Index: 92