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

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

Neuron 360 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.8005. Min Range: -4.8005

Max Act: 4.8005. Min Act: -0.0001

Data Index: 791186 (C4 (Web Text))

Max Activating Token Index: 106

Text #1

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.7859. Min Act: -0.0000

Data Index: 841663 (C4 (Web Text))

Max Activating Token Index: 99

Text #2

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.4724. Min Act: -0.0001

Data Index: 1334798 (C4 (Web Text))

Max Activating Token Index: 823

Text #3

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.6655. Min Act: -0.0001

Data Index: 306302 (C4 (Web Text))

Max Activating Token Index: 944

Text #4

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.3957. Min Act: -0.0000

Data Index: 1193523 (C4 (Web Text))

Max Activating Token Index: 715

Text #5

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.0605. Min Act: -0.0000

Data Index: 1165450 (C4 (Web Text))

Max Activating Token Index: 15

Text #6

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.1744. Min Act: -0.0001

Data Index: 1420693 (Python Code)

Max Activating Token Index: 929

Text #7

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.1961. Min Act: -0.0001

Data Index: 1490142 (Python Code)

Max Activating Token Index: 145

Text #8

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.8007. Min Act: -0.0001

Data Index: 1564013 (Python Code)

Max Activating Token Index: 357

Text #9

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.0633. Min Act: -0.0001

Data Index: 1567527 (Python Code)

Max Activating Token Index: 158

Text #10

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.8896. Min Act: -0.0001

Data Index: 476992 (C4 (Web Text))

Max Activating Token Index: 253

Text #11

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.8509. Min Act: -0.0001

Data Index: 152738 (C4 (Web Text))

Max Activating Token Index: 684

Text #12

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.7856. Min Act: -0.0001

Data Index: 1370789 (Python Code)

Max Activating Token Index: 164

Text #13

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.6768. Min Act: -0.0000

Data Index: 180338 (C4 (Web Text))

Max Activating Token Index: 787

Text #14

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.7252. Min Act: -0.0001

Data Index: 1253146 (C4 (Web Text))

Max Activating Token Index: 126

Text #15

Max Range: 4.8005. Min Range: -4.8005

Max Act: 3.0176. Min Act: -0.0001

Data Index: 1113355 (C4 (Web Text))

Max Activating Token Index: 907

Text #16

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.6958. Min Act: -0.0000

Data Index: 242129 (C4 (Web Text))

Max Activating Token Index: 390

Text #17

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.2644. Min Act: -0.0000

Data Index: 1254276 (C4 (Web Text))

Max Activating Token Index: 927

Text #18

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.6065. Min Act: -0.0000

Data Index: 1272957 (C4 (Web Text))

Max Activating Token Index: 962

Text #19

Max Range: 4.8005. Min Range: -4.8005

Max Act: 2.6210. Min Act: -0.0001

Data Index: 989938 (C4 (Web Text))

Max Activating Token Index: 106