Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1848 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.8901. Min Act: -0.0000

Data Index: 1152799 (C4 (Web Text))

Max Activating Token Index: 700

Text #1

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.9269. Min Act: -0.0000

Data Index: 784378 (C4 (Web Text))

Max Activating Token Index: 206

Text #2

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.8655. Min Act: -0.0000

Data Index: 685874 (C4 (Web Text))

Max Activating Token Index: 540

Text #3

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.7512. Min Act: -0.0000

Data Index: 146285 (C4 (Web Text))

Max Activating Token Index: 947

Text #4

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.9037. Min Act: -0.0000

Data Index: 1019604 (C4 (Web Text))

Max Activating Token Index: 815

Text #5

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.6925. Min Act: -0.0000

Data Index: 218252 (C4 (Web Text))

Max Activating Token Index: 587

Text #6

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.7450. Min Act: -0.0000

Data Index: 1360031 (Python Code)

Max Activating Token Index: 913

Text #7

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.7117. Min Act: -0.0000

Data Index: 1597703 (Python Code)

Max Activating Token Index: 611

Text #8

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.8241. Min Act: -0.0000

Data Index: 1277455 (C4 (Web Text))

Max Activating Token Index: 200

Text #9

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.5857. Min Act: -0.0000

Data Index: 1514760 (Python Code)

Max Activating Token Index: 152

Text #10

Max Range: 2.0114. Min Range: -2.0114

Max Act: 2.0114. Min Act: -0.0000

Data Index: 1120089 (C4 (Web Text))

Max Activating Token Index: 650

Text #11

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.6881. Min Act: 0.0000

Data Index: 1207676 (C4 (Web Text))

Max Activating Token Index: 957

Text #12

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.5428. Min Act: -0.0000

Data Index: 34149 (C4 (Web Text))

Max Activating Token Index: 686

Text #13

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.6078. Min Act: -0.0000

Data Index: 1023630 (C4 (Web Text))

Max Activating Token Index: 741

Text #14

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.7601. Min Act: -0.0000

Data Index: 884956 (C4 (Web Text))

Max Activating Token Index: 746

Text #15

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.6719. Min Act: -0.0000

Data Index: 884096 (C4 (Web Text))

Max Activating Token Index: 432

Text #16

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.5653. Min Act: -0.0000

Data Index: 17584 (C4 (Web Text))

Max Activating Token Index: 938

Text #17

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.5733. Min Act: -0.0000

Data Index: 1506437 (Python Code)

Max Activating Token Index: 576

Text #18

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.5447. Min Act: -0.0000

Data Index: 1634298 (Python Code)

Max Activating Token Index: 883

Text #19

Max Range: 2.0114. Min Range: -2.0114

Max Act: 1.4587. Min Act: -0.0000

Data Index: 1140364 (C4 (Web Text))

Max Activating Token Index: 77