Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4600 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 1.9195. Min Range: -1.9195

Max Act: 1.9195. Min Act: -0.0002

Data Index: 646081 (C4 (Web Text))

Max Activating Token Index: 142

Text #1

Max Range: 1.9195. Min Range: -1.9195

Max Act: 1.5084. Min Act: -0.0002

Data Index: 1245416 (C4 (Web Text))

Max Activating Token Index: 172

Text #2

Max Range: 1.9195. Min Range: -1.9195

Max Act: 1.1142. Min Act: -0.0001

Data Index: 1300051 (C4 (Web Text))

Max Activating Token Index: 33

Text #3

Max Range: 1.9195. Min Range: -1.9195

Max Act: 1.0957. Min Act: -0.0002

Data Index: 1399677 (Python Code)

Max Activating Token Index: 79

Text #4

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.8518. Min Act: -0.0002

Data Index: 1645093 (Python Code)

Max Activating Token Index: 22

Text #5

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.8081. Min Act: -0.0001

Data Index: 1619655 (Python Code)

Max Activating Token Index: 25

Text #6

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7694. Min Act: -0.0002

Data Index: 1407441 (Python Code)

Max Activating Token Index: 123

Text #7

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7708. Min Act: -0.0001

Data Index: 1578147 (Python Code)

Max Activating Token Index: 30

Text #8

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7302. Min Act: -0.0002

Data Index: 559169 (C4 (Web Text))

Max Activating Token Index: 24

Text #9

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7167. Min Act: -0.0002

Data Index: 522698 (C4 (Web Text))

Max Activating Token Index: 768

Text #10

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7195. Min Act: -0.0003

Data Index: 1128214 (C4 (Web Text))

Max Activating Token Index: 317

Text #11

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.7145. Min Act: -0.0002

Data Index: 1387732 (Python Code)

Max Activating Token Index: 201

Text #12

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6716. Min Act: -0.0002

Data Index: 1468473 (Python Code)

Max Activating Token Index: 93

Text #13

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6997. Min Act: -0.0002

Data Index: 1088010 (C4 (Web Text))

Max Activating Token Index: 58

Text #14

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6965. Min Act: -0.0002

Data Index: 1652526 (Python Code)

Max Activating Token Index: 23

Text #15

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6678. Min Act: -0.0002

Data Index: 1275955 (C4 (Web Text))

Max Activating Token Index: 304

Text #16

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6731. Min Act: -0.0001

Data Index: 1635357 (Python Code)

Max Activating Token Index: 93

Text #17

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6141. Min Act: -0.0002

Data Index: 844230 (C4 (Web Text))

Max Activating Token Index: 53

Text #18

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6023. Min Act: -0.0002

Data Index: 1276759 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 1.9195. Min Range: -1.9195

Max Act: 0.6039. Min Act: -0.0002

Data Index: 1341213 (C4 (Web Text))

Max Activating Token Index: 496