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

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

Neuron 2353 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.3518. Min Act: -0.0001

Data Index: 475961 (C4 (Web Text))

Max Activating Token Index: 8

Text #1

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.3142. Min Act: -0.0001

Data Index: 836729 (C4 (Web Text))

Max Activating Token Index: 400

Text #2

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.3748. Min Act: -0.0001

Data Index: 539142 (C4 (Web Text))

Max Activating Token Index: 906

Text #3

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.1961. Min Act: -0.0001

Data Index: 840180 (C4 (Web Text))

Max Activating Token Index: 412

Text #4

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.1997. Min Act: -0.0001

Data Index: 1218573 (C4 (Web Text))

Max Activating Token Index: 900

Text #5

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.9882. Min Act: -0.0001

Data Index: 852291 (C4 (Web Text))

Max Activating Token Index: 198

Text #6

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.1248. Min Act: -0.0001

Data Index: 889778 (C4 (Web Text))

Max Activating Token Index: 406

Text #7

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.1145. Min Act: -0.0001

Data Index: 873542 (C4 (Web Text))

Max Activating Token Index: 1005

Text #8

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.2320. Min Act: -0.0001

Data Index: 882192 (C4 (Web Text))

Max Activating Token Index: 660

Text #9

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.0990. Min Act: -0.0001

Data Index: 757900 (C4 (Web Text))

Max Activating Token Index: 849

Text #10

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.1759. Min Act: -0.0001

Data Index: 764819 (C4 (Web Text))

Max Activating Token Index: 962

Text #11

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.9163. Min Act: -0.0001

Data Index: 5445 (C4 (Web Text))

Max Activating Token Index: 846

Text #12

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.7318. Min Act: -0.0001

Data Index: 355104 (C4 (Web Text))

Max Activating Token Index: 737

Text #13

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.9870. Min Act: -0.0001

Data Index: 421211 (C4 (Web Text))

Max Activating Token Index: 779

Text #14

Max Range: 3.3748. Min Range: -3.3748

Max Act: 3.0866. Min Act: -0.0001

Data Index: 969701 (C4 (Web Text))

Max Activating Token Index: 877

Text #15

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.8115. Min Act: -0.0001

Data Index: 1377415 (Python Code)

Max Activating Token Index: 238

Text #16

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.9003. Min Act: -0.0001

Data Index: 325663 (C4 (Web Text))

Max Activating Token Index: 390

Text #17

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.6587. Min Act: -0.0001

Data Index: 626135 (C4 (Web Text))

Max Activating Token Index: 965

Text #18

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.8664. Min Act: -0.0000

Data Index: 1647004 (Python Code)

Max Activating Token Index: 159

Text #19

Max Range: 3.3748. Min Range: -3.3748

Max Act: 2.7128. Min Act: -0.0001

Data Index: 749075 (C4 (Web Text))

Max Activating Token Index: 12