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

Dataset: The Pile

Neuron 3823 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.2587. Min Act: -0.0000

Data Index: 947803 (The Pile)

Max Activating Token Index: 246

Text #1

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.1132. Min Act: -0.0000

Data Index: 953824 (The Pile)

Max Activating Token Index: 72

Text #2

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.3955. Min Act: -0.0000

Data Index: 1555595 (The Pile)

Max Activating Token Index: 157

Text #3

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.2470. Min Act: -0.0000

Data Index: 118932 (The Pile)

Max Activating Token Index: 214

Text #4

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.1184. Min Act: -0.0000

Data Index: 227878 (The Pile)

Max Activating Token Index: 67

Text #5

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.0564. Min Act: -0.0000

Data Index: 1476905 (The Pile)

Max Activating Token Index: 203

Text #6

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.0959. Min Act: -0.0000

Data Index: 655072 (The Pile)

Max Activating Token Index: 74

Text #7

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.1022. Min Act: -0.0000

Data Index: 1346092 (The Pile)

Max Activating Token Index: 96

Text #8

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.9016. Min Act: -0.0000

Data Index: 1601360 (The Pile)

Max Activating Token Index: 230

Text #9

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.0844. Min Act: -0.0000

Data Index: 731556 (The Pile)

Max Activating Token Index: 83

Text #10

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.0583. Min Act: -0.0000

Data Index: 81313 (The Pile)

Max Activating Token Index: 246

Text #11

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.2532. Min Act: -0.0000

Data Index: 369812 (The Pile)

Max Activating Token Index: 186

Text #12

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.2194. Min Act: -0.0000

Data Index: 1285786 (The Pile)

Max Activating Token Index: 125

Text #13

Max Range: 8.3955. Min Range: -8.3955

Max Act: 8.0069. Min Act: -0.0000

Data Index: 1458937 (The Pile)

Max Activating Token Index: 158

Text #14

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.9993. Min Act: -0.0000

Data Index: 1383735 (The Pile)

Max Activating Token Index: 107

Text #15

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.9987. Min Act: -0.0000

Data Index: 1307796 (The Pile)

Max Activating Token Index: 109

Text #16

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.8763. Min Act: -0.0000

Data Index: 1645214 (The Pile)

Max Activating Token Index: 84

Text #17

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.7762. Min Act: -0.0000

Data Index: 179565 (The Pile)

Max Activating Token Index: 109

Text #18

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.9375. Min Act: -0.0000

Data Index: 1800134 (The Pile)

Max Activating Token Index: 103

Text #19

Max Range: 8.3955. Min Range: -8.3955

Max Act: 7.8854. Min Act: -0.0000

Data Index: 1042268 (The Pile)

Max Activating Token Index: 134