Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 901 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 1.4050. Min Range: -1.4050

Max Act: 1.4050. Min Act: -0.0000

Data Index: 1847981 (The Pile)

Max Activating Token Index: 425

Text #1

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8124. Min Act: -0.0000

Data Index: 1431863 (The Pile)

Max Activating Token Index: 119

Text #2

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8277. Min Act: -0.0000

Data Index: 359831 (The Pile)

Max Activating Token Index: 35

Text #3

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.9088. Min Act: -0.0000

Data Index: 1942138 (The Pile)

Max Activating Token Index: 10

Text #4

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8142. Min Act: -0.0000

Data Index: 262325 (The Pile)

Max Activating Token Index: 337

Text #5

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7164. Min Act: -0.0000

Data Index: 968509 (The Pile)

Max Activating Token Index: 462

Text #6

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8977. Min Act: -0.0000

Data Index: 1268003 (The Pile)

Max Activating Token Index: 621

Text #7

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8351. Min Act: -0.0000

Data Index: 386358 (The Pile)

Max Activating Token Index: 490

Text #8

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.8505. Min Act: -0.0000

Data Index: 1598969 (The Pile)

Max Activating Token Index: 772

Text #9

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7784. Min Act: -0.0000

Data Index: 89019 (The Pile)

Max Activating Token Index: 372

Text #10

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.6850. Min Act: -0.0000

Data Index: 1331314 (The Pile)

Max Activating Token Index: 439

Text #11

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7682. Min Act: -0.0000

Data Index: 964862 (The Pile)

Max Activating Token Index: 68

Text #12

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7663. Min Act: -0.0000

Data Index: 61225 (The Pile)

Max Activating Token Index: 567

Text #13

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7134. Min Act: -0.0000

Data Index: 1558483 (The Pile)

Max Activating Token Index: 68

Text #14

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7157. Min Act: -0.0000

Data Index: 1413078 (The Pile)

Max Activating Token Index: 162

Text #15

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.7316. Min Act: -0.0000

Data Index: 1164685 (The Pile)

Max Activating Token Index: 753

Text #16

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.6614. Min Act: -0.0000

Data Index: 1273698 (The Pile)

Max Activating Token Index: 270

Text #17

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.5988. Min Act: -0.0000

Data Index: 539528 (The Pile)

Max Activating Token Index: 410

Text #18

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.4246. Min Act: -0.0000

Data Index: 1606010 (The Pile)

Max Activating Token Index: 943

Text #19

Max Range: 1.4050. Min Range: -1.4050

Max Act: 0.6611. Min Act: -0.0000

Data Index: 1955375 (The Pile)

Max Activating Token Index: 8