Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1820 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.0795. Min Range: -1.0795

Max Act: 1.0795. Min Act: -0.0001

Data Index: 1550277 (Python Code)

Max Activating Token Index: 1015

Text #1

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9574. Min Act: 0.0001

Data Index: 434160 (C4 (Web Text))

Max Activating Token Index: 127

Text #2

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9430. Min Act: 0.0001

Data Index: 109738 (C4 (Web Text))

Max Activating Token Index: 117

Text #3

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9435. Min Act: 0.0002

Data Index: 729141 (C4 (Web Text))

Max Activating Token Index: 76

Text #4

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9269. Min Act: 0.0001

Data Index: 317341 (C4 (Web Text))

Max Activating Token Index: 62

Text #5

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9006. Min Act: 0.0001

Data Index: 460485 (C4 (Web Text))

Max Activating Token Index: 253

Text #6

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9023. Min Act: 0.0002

Data Index: 1020739 (C4 (Web Text))

Max Activating Token Index: 193

Text #7

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.9123. Min Act: 0.0001

Data Index: 1279150 (C4 (Web Text))

Max Activating Token Index: 65

Text #8

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8632. Min Act: 0.0001

Data Index: 485616 (C4 (Web Text))

Max Activating Token Index: 93

Text #9

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8245. Min Act: -0.0000

Data Index: 66682 (C4 (Web Text))

Max Activating Token Index: 608

Text #10

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8459. Min Act: 0.0001

Data Index: 1178214 (C4 (Web Text))

Max Activating Token Index: 305

Text #11

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8597. Min Act: 0.0001

Data Index: 1013983 (C4 (Web Text))

Max Activating Token Index: 161

Text #12

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8275. Min Act: 0.0001

Data Index: 96988 (C4 (Web Text))

Max Activating Token Index: 63

Text #13

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8252. Min Act: 0.0001

Data Index: 973999 (C4 (Web Text))

Max Activating Token Index: 65

Text #14

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8176. Min Act: -0.0000

Data Index: 351176 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8476. Min Act: -0.0001

Data Index: 1428893 (Python Code)

Max Activating Token Index: 244

Text #16

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8170. Min Act: 0.0001

Data Index: 389103 (C4 (Web Text))

Max Activating Token Index: 97

Text #17

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8188. Min Act: 0.0000

Data Index: 1346066 (C4 (Web Text))

Max Activating Token Index: 85

Text #18

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8142. Min Act: 0.0000

Data Index: 525175 (C4 (Web Text))

Max Activating Token Index: 36

Text #19

Max Range: 1.0795. Min Range: -1.0795

Max Act: 0.8027. Min Act: 0.0001

Data Index: 933827 (C4 (Web Text))

Max Activating Token Index: 125