Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 316 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.8574. Min Act: -0.0005

Data Index: 1577207 (Python Code)

Max Activating Token Index: 25

Text #1

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.8036. Min Act: -0.0005

Data Index: 1521558 (Python Code)

Max Activating Token Index: 12

Text #2

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.7269. Min Act: -0.0005

Data Index: 1394312 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.7169. Min Act: -0.0005

Data Index: 1467037 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6846. Min Act: -0.0005

Data Index: 1099392 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6825. Min Act: -0.0005

Data Index: 1646290 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6941. Min Act: -0.0005

Data Index: 1288819 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6423. Min Act: -0.0005

Data Index: 1555689 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6418. Min Act: -0.0005

Data Index: 1510640 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6696. Min Act: -0.0005

Data Index: 1604156 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6868. Min Act: -0.0005

Data Index: 1606759 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6630. Min Act: -0.0005

Data Index: 1574881 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6192. Min Act: -0.0005

Data Index: 1433547 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6842. Min Act: -0.0005

Data Index: 1601409 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6336. Min Act: -0.0005

Data Index: 1623784 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6125. Min Act: -0.0005

Data Index: 1439391 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6674. Min Act: -0.0005

Data Index: 1434077 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6034. Min Act: -0.0005

Data Index: 1257439 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6378. Min Act: -0.0005

Data Index: 1093326 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.8574. Min Range: -0.8574

Max Act: 0.6542. Min Act: -0.0005

Data Index: 282348 (C4 (Web Text))

Max Activating Token Index: 903