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

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

Neuron 157 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0102. Min Act: -0.0006

Data Index: 458827 (C4 (Web Text))

Max Activating Token Index: 985

Text #1

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0102. Min Act: -0.0005

Data Index: 1571283 (Python Code)

Max Activating Token Index: 284

Text #2

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0099. Min Act: -0.0005

Data Index: 1335229 (C4 (Web Text))

Max Activating Token Index: 307

Text #3

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0097. Min Act: -0.0005

Data Index: 1494809 (Python Code)

Max Activating Token Index: 925

Text #4

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0093. Min Act: -0.0005

Data Index: 373184 (C4 (Web Text))

Max Activating Token Index: 768

Text #5

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0093. Min Act: -0.0005

Data Index: 1613636 (Python Code)

Max Activating Token Index: 923

Text #6

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0093. Min Act: -0.0005

Data Index: 274905 (C4 (Web Text))

Max Activating Token Index: 96

Text #7

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0093. Min Act: -0.0005

Data Index: 461424 (C4 (Web Text))

Max Activating Token Index: 947

Text #8

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0091. Min Act: -0.0006

Data Index: 453278 (C4 (Web Text))

Max Activating Token Index: 719

Text #9

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0090. Min Act: -0.0005

Data Index: 679545 (C4 (Web Text))

Max Activating Token Index: 307

Text #10

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0089. Min Act: -0.0005

Data Index: 1098918 (C4 (Web Text))

Max Activating Token Index: 584

Text #11

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0090. Min Act: -0.0005

Data Index: 469326 (C4 (Web Text))

Max Activating Token Index: 16

Text #12

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0089. Min Act: -0.0005

Data Index: 59927 (C4 (Web Text))

Max Activating Token Index: 591

Text #13

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0089. Min Act: -0.0005

Data Index: 1367054 (Python Code)

Max Activating Token Index: 594

Text #14

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0089. Min Act: -0.0005

Data Index: 446804 (C4 (Web Text))

Max Activating Token Index: 883

Text #15

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0090. Min Act: -0.0006

Data Index: 464866 (C4 (Web Text))

Max Activating Token Index: 334

Text #16

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0089. Min Act: -0.0005

Data Index: 1432514 (Python Code)

Max Activating Token Index: 196

Text #17

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0088. Min Act: -0.0005

Data Index: 898563 (C4 (Web Text))

Max Activating Token Index: 665

Text #18

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0088. Min Act: -0.0005

Data Index: 642460 (C4 (Web Text))

Max Activating Token Index: 48

Text #19

Max Range: 0.0102. Min Range: -0.0102

Max Act: 0.0088. Min Act: -0.0005

Data Index: 791008 (C4 (Web Text))

Max Activating Token Index: 625