Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3518 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5936. Min Act: -0.0001

Data Index: 918805 (C4 (Web Text))

Max Activating Token Index: 323

Text #1

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.6105. Min Act: -0.0001

Data Index: 1284563 (C4 (Web Text))

Max Activating Token Index: 610

Text #2

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5807. Min Act: -0.0001

Data Index: 989805 (C4 (Web Text))

Max Activating Token Index: 31

Text #3

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5588. Min Act: -0.0001

Data Index: 310104 (C4 (Web Text))

Max Activating Token Index: 528

Text #4

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5170. Min Act: -0.0001

Data Index: 998318 (C4 (Web Text))

Max Activating Token Index: 529

Text #5

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.4823. Min Act: -0.0001

Data Index: 1370139 (Python Code)

Max Activating Token Index: 586

Text #6

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.4874. Min Act: -0.0002

Data Index: 47591 (C4 (Web Text))

Max Activating Token Index: 724

Text #7

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5685. Min Act: -0.0001

Data Index: 224849 (C4 (Web Text))

Max Activating Token Index: 208

Text #8

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5482. Min Act: -0.0001

Data Index: 631318 (C4 (Web Text))

Max Activating Token Index: 67

Text #9

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5416. Min Act: -0.0001

Data Index: 1114085 (C4 (Web Text))

Max Activating Token Index: 58

Text #10

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.3964. Min Act: -0.0001

Data Index: 1260497 (C4 (Web Text))

Max Activating Token Index: 249

Text #11

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.4520. Min Act: -0.0001

Data Index: 474759 (C4 (Web Text))

Max Activating Token Index: 74

Text #12

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.4902. Min Act: -0.0001

Data Index: 821280 (C4 (Web Text))

Max Activating Token Index: 312

Text #13

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5449. Min Act: -0.0001

Data Index: 326497 (C4 (Web Text))

Max Activating Token Index: 897

Text #14

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5735. Min Act: -0.0001

Data Index: 864089 (C4 (Web Text))

Max Activating Token Index: 558

Text #15

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5554. Min Act: -0.0001

Data Index: 398279 (C4 (Web Text))

Max Activating Token Index: 653

Text #16

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.5169. Min Act: -0.0002

Data Index: 1469431 (Python Code)

Max Activating Token Index: 1012

Text #17

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.3461. Min Act: -0.0001

Data Index: 627238 (C4 (Web Text))

Max Activating Token Index: 721

Text #18

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.3988. Min Act: -0.0001

Data Index: 1408718 (Python Code)

Max Activating Token Index: 885

Text #19

Max Range: 0.6105. Min Range: -0.6105

Max Act: 0.4554. Min Act: -0.0001

Data Index: 1097298 (C4 (Web Text))

Max Activating Token Index: 609