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

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

Neuron 3161 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3971. Min Range: -3.3971

Max Act: 3.0687. Min Act: -0.0000

Data Index: 772556 (C4 (Web Text))

Max Activating Token Index: 608

Text #1

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.9148. Min Act: -0.0000

Data Index: 1277654 (C4 (Web Text))

Max Activating Token Index: 400

Text #2

Max Range: 3.3971. Min Range: -3.3971

Max Act: 3.3971. Min Act: -0.0000

Data Index: 132858 (C4 (Web Text))

Max Activating Token Index: 832

Text #3

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.7977. Min Act: -0.0000

Data Index: 1382652 (Python Code)

Max Activating Token Index: 250

Text #4

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.8138. Min Act: -0.0000

Data Index: 123542 (C4 (Web Text))

Max Activating Token Index: 920

Text #5

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.4734. Min Act: -0.0000

Data Index: 775183 (C4 (Web Text))

Max Activating Token Index: 607

Text #6

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.0775. Min Act: -0.0000

Data Index: 923388 (C4 (Web Text))

Max Activating Token Index: 240

Text #7

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.0489. Min Act: -0.0000

Data Index: 310683 (C4 (Web Text))

Max Activating Token Index: 630

Text #8

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.3280. Min Act: -0.0000

Data Index: 1593982 (Python Code)

Max Activating Token Index: 885

Text #9

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.1597. Min Act: -0.0000

Data Index: 1613976 (Python Code)

Max Activating Token Index: 536

Text #10

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.7627. Min Act: -0.0000

Data Index: 586427 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.0525. Min Act: -0.0000

Data Index: 1018023 (C4 (Web Text))

Max Activating Token Index: 186

Text #12

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.0530. Min Act: -0.0000

Data Index: 746692 (C4 (Web Text))

Max Activating Token Index: 79

Text #13

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.5446. Min Act: -0.0000

Data Index: 1552069 (Python Code)

Max Activating Token Index: 968

Text #14

Max Range: 3.3971. Min Range: -3.3971

Max Act: 2.0286. Min Act: -0.0000

Data Index: 1246181 (C4 (Web Text))

Max Activating Token Index: 613

Text #15

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.7258. Min Act: -0.0000

Data Index: 1627193 (Python Code)

Max Activating Token Index: 799

Text #16

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.8036. Min Act: -0.0000

Data Index: 1597204 (Python Code)

Max Activating Token Index: 766

Text #17

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.9817. Min Act: -0.0000

Data Index: 1518609 (Python Code)

Max Activating Token Index: 949

Text #18

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.4515. Min Act: -0.0000

Data Index: 1459490 (Python Code)

Max Activating Token Index: 595

Text #19

Max Range: 3.3971. Min Range: -3.3971

Max Act: 1.5232. Min Act: -0.0000

Data Index: 486627 (C4 (Web Text))

Max Activating Token Index: 672