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

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

Neuron 1215 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3375. Min Act: -0.0002

Data Index: 243201 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3271. Min Act: -0.0002

Data Index: 923315 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3146. Min Act: -0.0002

Data Index: 1390110 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3260. Min Act: -0.0002

Data Index: 1554322 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2922. Min Act: -0.0002

Data Index: 541440 (C4 (Web Text))

Max Activating Token Index: 441

Text #5

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3139. Min Act: -0.0002

Data Index: 1637225 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3236. Min Act: -0.0002

Data Index: 1368886 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3126. Min Act: -0.0002

Data Index: 612422 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2955. Min Act: -0.0002

Data Index: 335531 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3071. Min Act: -0.0002

Data Index: 1535349 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.3003. Min Act: -0.0002

Data Index: 1368854 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2781. Min Act: -0.0002

Data Index: 719311 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2962. Min Act: -0.0002

Data Index: 1647722 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2719. Min Act: -0.0002

Data Index: 1096834 (C4 (Web Text))

Max Activating Token Index: 58

Text #14

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2738. Min Act: -0.0002

Data Index: 1263036 (C4 (Web Text))

Max Activating Token Index: 938

Text #15

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2749. Min Act: -0.0002

Data Index: 1505745 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2946. Min Act: -0.0002

Data Index: 1453746 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2808. Min Act: -0.0002

Data Index: 1595682 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2783. Min Act: -0.0002

Data Index: 1176788 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.3375. Min Range: -0.3375

Max Act: 0.2887. Min Act: -0.0002

Data Index: 1484682 (Python Code)

Max Activating Token Index: 1023