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

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

Neuron 1078 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.9955. Min Range: -8.9955

Max Act: 8.0336. Min Act: -0.0000

Data Index: 385274 (C4 (Web Text))

Max Activating Token Index: 74

Text #1

Max Range: 8.9955. Min Range: -8.9955

Max Act: 8.9955. Min Act: -0.0000

Data Index: 209495 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 8.9955. Min Range: -8.9955

Max Act: 7.5675. Min Act: -0.0000

Data Index: 240521 (C4 (Web Text))

Max Activating Token Index: 30

Text #3

Max Range: 8.9955. Min Range: -8.9955

Max Act: 7.4181. Min Act: -0.0000

Data Index: 121154 (C4 (Web Text))

Max Activating Token Index: 19

Text #4

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9405. Min Act: -0.0000

Data Index: 1328885 (C4 (Web Text))

Max Activating Token Index: 275

Text #5

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9098. Min Act: -0.0000

Data Index: 1345424 (C4 (Web Text))

Max Activating Token Index: 22

Text #6

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.8934. Min Act: -0.0000

Data Index: 521578 (C4 (Web Text))

Max Activating Token Index: 71

Text #7

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9905. Min Act: -0.0000

Data Index: 651184 (C4 (Web Text))

Max Activating Token Index: 119

Text #8

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9665. Min Act: -0.0000

Data Index: 253687 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9075. Min Act: -0.0000

Data Index: 113536 (C4 (Web Text))

Max Activating Token Index: 124

Text #10

Max Range: 8.9955. Min Range: -8.9955

Max Act: 7.0962. Min Act: -0.0000

Data Index: 726090 (C4 (Web Text))

Max Activating Token Index: 172

Text #11

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.9236. Min Act: -0.0000

Data Index: 1648206 (Python Code)

Max Activating Token Index: 142

Text #12

Max Range: 8.9955. Min Range: -8.9955

Max Act: 7.1878. Min Act: -0.0000

Data Index: 228533 (C4 (Web Text))

Max Activating Token Index: 644

Text #13

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.5647. Min Act: -0.0000

Data Index: 1207828 (C4 (Web Text))

Max Activating Token Index: 303

Text #14

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.7253. Min Act: -0.0000

Data Index: 608649 (C4 (Web Text))

Max Activating Token Index: 749

Text #15

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.0973. Min Act: -0.0000

Data Index: 1476463 (Python Code)

Max Activating Token Index: 350

Text #16

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.8697. Min Act: -0.0000

Data Index: 92813 (C4 (Web Text))

Max Activating Token Index: 494

Text #17

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.4184. Min Act: -0.0000

Data Index: 615608 (C4 (Web Text))

Max Activating Token Index: 373

Text #18

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.6148. Min Act: -0.0000

Data Index: 339237 (C4 (Web Text))

Max Activating Token Index: 48

Text #19

Max Range: 8.9955. Min Range: -8.9955

Max Act: 6.8125. Min Act: -0.0000

Data Index: 443493 (C4 (Web Text))

Max Activating Token Index: 42