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

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

Neuron 454 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0092. Min Act: -0.0006

Data Index: 748667 (C4 (Web Text))

Max Activating Token Index: 922

Text #1

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0092. Min Act: -0.0006

Data Index: 272522 (C4 (Web Text))

Max Activating Token Index: 814

Text #2

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0082. Min Act: -0.0006

Data Index: 942129 (C4 (Web Text))

Max Activating Token Index: 732

Text #3

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0085. Min Act: -0.0006

Data Index: 766241 (C4 (Web Text))

Max Activating Token Index: 784

Text #4

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0081. Min Act: -0.0006

Data Index: 186203 (C4 (Web Text))

Max Activating Token Index: 712

Text #5

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0079. Min Act: -0.0006

Data Index: 913483 (C4 (Web Text))

Max Activating Token Index: 916

Text #6

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0084. Min Act: -0.0006

Data Index: 1341848 (C4 (Web Text))

Max Activating Token Index: 1018

Text #7

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0078. Min Act: -0.0006

Data Index: 523809 (C4 (Web Text))

Max Activating Token Index: 898

Text #8

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0071. Min Act: -0.0006

Data Index: 548542 (C4 (Web Text))

Max Activating Token Index: 608

Text #9

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0073. Min Act: -0.0006

Data Index: 557274 (C4 (Web Text))

Max Activating Token Index: 796

Text #10

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0074. Min Act: -0.0006

Data Index: 357457 (C4 (Web Text))

Max Activating Token Index: 395

Text #11

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0072. Min Act: -0.0006

Data Index: 1292502 (C4 (Web Text))

Max Activating Token Index: 154

Text #12

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0073. Min Act: -0.0006

Data Index: 1256629 (C4 (Web Text))

Max Activating Token Index: 122

Text #13

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0070. Min Act: -0.0006

Data Index: 1526802 (Python Code)

Max Activating Token Index: 195

Text #14

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0074. Min Act: -0.0006

Data Index: 186797 (C4 (Web Text))

Max Activating Token Index: 723

Text #15

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0070. Min Act: -0.0006

Data Index: 1311004 (C4 (Web Text))

Max Activating Token Index: 1000

Text #16

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0064. Min Act: -0.0006

Data Index: 239898 (C4 (Web Text))

Max Activating Token Index: 326

Text #17

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0068. Min Act: -0.0006

Data Index: 608198 (C4 (Web Text))

Max Activating Token Index: 188

Text #18

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0068. Min Act: -0.0006

Data Index: 213886 (C4 (Web Text))

Max Activating Token Index: 690

Text #19

Max Range: 0.0092. Min Range: -0.0092

Max Act: 0.0068. Min Act: -0.0006

Data Index: 821438 (C4 (Web Text))

Max Activating Token Index: 803