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

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

Neuron 1696 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1692. Min Act: -0.0000

Data Index: 698069 (C4 (Web Text))

Max Activating Token Index: 102

Text #1

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1740. Min Act: -0.0000

Data Index: 368356 (C4 (Web Text))

Max Activating Token Index: 288

Text #2

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1545. Min Act: -0.0000

Data Index: 1140455 (C4 (Web Text))

Max Activating Token Index: 439

Text #3

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1429. Min Act: -0.0000

Data Index: 884988 (C4 (Web Text))

Max Activating Token Index: 440

Text #4

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1396. Min Act: -0.0000

Data Index: 450983 (C4 (Web Text))

Max Activating Token Index: 884

Text #5

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1542. Min Act: -0.0000

Data Index: 984981 (C4 (Web Text))

Max Activating Token Index: 932

Text #6

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1402. Min Act: -0.0000

Data Index: 512723 (C4 (Web Text))

Max Activating Token Index: 909

Text #7

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1259. Min Act: -0.0000

Data Index: 977390 (C4 (Web Text))

Max Activating Token Index: 771

Text #8

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1274. Min Act: -0.0000

Data Index: 1004889 (C4 (Web Text))

Max Activating Token Index: 27

Text #9

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1282. Min Act: -0.0000

Data Index: 978758 (C4 (Web Text))

Max Activating Token Index: 197

Text #10

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1166. Min Act: -0.0000

Data Index: 1135280 (C4 (Web Text))

Max Activating Token Index: 1004

Text #11

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1172. Min Act: -0.0000

Data Index: 1531894 (Python Code)

Max Activating Token Index: 365

Text #12

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1132. Min Act: -0.0000

Data Index: 1547850 (Python Code)

Max Activating Token Index: 48

Text #13

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1080. Min Act: -0.0000

Data Index: 302783 (C4 (Web Text))

Max Activating Token Index: 922

Text #14

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.0998. Min Act: -0.0000

Data Index: 1253347 (C4 (Web Text))

Max Activating Token Index: 134

Text #15

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1098. Min Act: -0.0000

Data Index: 1643953 (Python Code)

Max Activating Token Index: 461

Text #16

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1130. Min Act: -0.0000

Data Index: 1537745 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1028. Min Act: -0.0000

Data Index: 1592630 (Python Code)

Max Activating Token Index: 284

Text #18

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.1052. Min Act: -0.0000

Data Index: 367242 (C4 (Web Text))

Max Activating Token Index: 503

Text #19

Max Range: 0.1740. Min Range: -0.1740

Max Act: 0.0977. Min Act: -0.0000

Data Index: 1174710 (C4 (Web Text))

Max Activating Token Index: 781