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

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

Neuron 3790 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.3239. Min Act: -0.0000

Data Index: 733589 (C4 (Web Text))

Max Activating Token Index: 994

Text #1

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2869. Min Act: -0.0000

Data Index: 735872 (C4 (Web Text))

Max Activating Token Index: 640

Text #2

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2896. Min Act: -0.0000

Data Index: 1085873 (C4 (Web Text))

Max Activating Token Index: 939

Text #3

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2894. Min Act: -0.0000

Data Index: 763085 (C4 (Web Text))

Max Activating Token Index: 17

Text #4

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2830. Min Act: -0.0000

Data Index: 681719 (C4 (Web Text))

Max Activating Token Index: 159

Text #5

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.3050. Min Act: -0.0000

Data Index: 346471 (C4 (Web Text))

Max Activating Token Index: 440

Text #6

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2721. Min Act: -0.0000

Data Index: 1204808 (C4 (Web Text))

Max Activating Token Index: 646

Text #7

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2792. Min Act: -0.0000

Data Index: 33428 (C4 (Web Text))

Max Activating Token Index: 22

Text #8

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2617. Min Act: -0.0000

Data Index: 1143872 (C4 (Web Text))

Max Activating Token Index: 846

Text #9

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2411. Min Act: -0.0000

Data Index: 1621773 (Python Code)

Max Activating Token Index: 143

Text #10

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2228. Min Act: -0.0000

Data Index: 1434693 (Python Code)

Max Activating Token Index: 320

Text #11

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2318. Min Act: -0.0000

Data Index: 1237594 (C4 (Web Text))

Max Activating Token Index: 269

Text #12

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2121. Min Act: -0.0000

Data Index: 1378907 (Python Code)

Max Activating Token Index: 533

Text #13

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2196. Min Act: -0.0000

Data Index: 1044650 (C4 (Web Text))

Max Activating Token Index: 134

Text #14

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2073. Min Act: -0.0000

Data Index: 1373214 (Python Code)

Max Activating Token Index: 179

Text #15

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2186. Min Act: -0.0000

Data Index: 36719 (C4 (Web Text))

Max Activating Token Index: 356

Text #16

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2131. Min Act: -0.0000

Data Index: 909128 (C4 (Web Text))

Max Activating Token Index: 144

Text #17

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2187. Min Act: -0.0000

Data Index: 1574485 (Python Code)

Max Activating Token Index: 764

Text #18

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.2239. Min Act: -0.0001

Data Index: 1476945 (Python Code)

Max Activating Token Index: 936

Text #19

Max Range: 0.3239. Min Range: -0.3239

Max Act: 0.1992. Min Act: -0.0000

Data Index: 1367046 (Python Code)

Max Activating Token Index: 393