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

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

Neuron 1635 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: 0.9366. Min Range: -0.9366

Max Act: 0.9366. Min Act: -0.0001

Data Index: 824917 (C4 (Web Text))

Max Activating Token Index: 746

Text #1

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.8846. Min Act: -0.0001

Data Index: 684635 (C4 (Web Text))

Max Activating Token Index: 87

Text #2

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7887. Min Act: -0.0001

Data Index: 869771 (C4 (Web Text))

Max Activating Token Index: 74

Text #3

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.8001. Min Act: -0.0001

Data Index: 415774 (C4 (Web Text))

Max Activating Token Index: 955

Text #4

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.8482. Min Act: -0.0001

Data Index: 208677 (C4 (Web Text))

Max Activating Token Index: 126

Text #5

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7495. Min Act: -0.0001

Data Index: 630866 (C4 (Web Text))

Max Activating Token Index: 841

Text #6

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7182. Min Act: -0.0001

Data Index: 183159 (C4 (Web Text))

Max Activating Token Index: 824

Text #7

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7530. Min Act: -0.0001

Data Index: 421316 (C4 (Web Text))

Max Activating Token Index: 727

Text #8

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7055. Min Act: -0.0001

Data Index: 596071 (C4 (Web Text))

Max Activating Token Index: 868

Text #9

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.8049. Min Act: -0.0001

Data Index: 591009 (C4 (Web Text))

Max Activating Token Index: 834

Text #10

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7826. Min Act: -0.0001

Data Index: 67712 (C4 (Web Text))

Max Activating Token Index: 139

Text #11

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7817. Min Act: -0.0001

Data Index: 897719 (C4 (Web Text))

Max Activating Token Index: 953

Text #12

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.6951. Min Act: -0.0001

Data Index: 952691 (C4 (Web Text))

Max Activating Token Index: 324

Text #13

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7449. Min Act: -0.0001

Data Index: 711020 (C4 (Web Text))

Max Activating Token Index: 389

Text #14

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7033. Min Act: -0.0001

Data Index: 1325372 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7329. Min Act: -0.0001

Data Index: 1565723 (Python Code)

Max Activating Token Index: 217

Text #16

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7603. Min Act: -0.0001

Data Index: 243725 (C4 (Web Text))

Max Activating Token Index: 988

Text #17

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7162. Min Act: -0.0001

Data Index: 613976 (C4 (Web Text))

Max Activating Token Index: 788

Text #18

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7140. Min Act: -0.0001

Data Index: 365022 (C4 (Web Text))

Max Activating Token Index: 855

Text #19

Max Range: 0.9366. Min Range: -0.9366

Max Act: 0.7597. Min Act: -0.0001

Data Index: 116381 (C4 (Web Text))

Max Activating Token Index: 766