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

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

Neuron 4022 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1081. Min Range: -2.1081

Max Act: 2.1081. Min Act: -0.0000

Data Index: 1226523 (C4 (Web Text))

Max Activating Token Index: 644

Text #1

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.6780. Min Act: -0.0000

Data Index: 50410 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.7209. Min Act: -0.0000

Data Index: 1517231 (Python Code)

Max Activating Token Index: 979

Text #3

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.6239. Min Act: -0.0000

Data Index: 674364 (C4 (Web Text))

Max Activating Token Index: 1019

Text #4

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.6300. Min Act: -0.0000

Data Index: 1048606 (C4 (Web Text))

Max Activating Token Index: 301

Text #5

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.7136. Min Act: -0.0000

Data Index: 1607644 (Python Code)

Max Activating Token Index: 1019

Text #6

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.5726. Min Act: -0.0000

Data Index: 281553 (C4 (Web Text))

Max Activating Token Index: 391

Text #7

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.5341. Min Act: -0.0000

Data Index: 1471484 (Python Code)

Max Activating Token Index: 110

Text #8

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.1302. Min Act: -0.0000

Data Index: 1298579 (C4 (Web Text))

Max Activating Token Index: 749

Text #9

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.2669. Min Act: -0.0000

Data Index: 497679 (C4 (Web Text))

Max Activating Token Index: 338

Text #10

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.6350. Min Act: -0.0000

Data Index: 1405700 (Python Code)

Max Activating Token Index: 925

Text #11

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.5305. Min Act: -0.0000

Data Index: 1570476 (Python Code)

Max Activating Token Index: 28

Text #12

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.4712. Min Act: -0.0000

Data Index: 9811 (C4 (Web Text))

Max Activating Token Index: 44

Text #13

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.5142. Min Act: -0.0000

Data Index: 1367350 (Python Code)

Max Activating Token Index: 147

Text #14

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.5157. Min Act: -0.0000

Data Index: 891982 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.2305. Min Act: -0.0000

Data Index: 512436 (C4 (Web Text))

Max Activating Token Index: 151

Text #16

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.4985. Min Act: -0.0000

Data Index: 1552583 (Python Code)

Max Activating Token Index: 488

Text #17

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.4436. Min Act: -0.0000

Data Index: 1434862 (Python Code)

Max Activating Token Index: 130

Text #18

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.2194. Min Act: -0.0000

Data Index: 1246964 (C4 (Web Text))

Max Activating Token Index: 380

Text #19

Max Range: 2.1081. Min Range: -2.1081

Max Act: 1.4577. Min Act: -0.0000

Data Index: 1489116 (Python Code)

Max Activating Token Index: 983