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

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

Neuron 3319 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.8236. Min Act: -0.0001

Data Index: 1301848 (C4 (Web Text))

Max Activating Token Index: 127

Text #1

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.8683. Min Act: -0.0001

Data Index: 108908 (C4 (Web Text))

Max Activating Token Index: 63

Text #2

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7889. Min Act: -0.0001

Data Index: 31859 (C4 (Web Text))

Max Activating Token Index: 772

Text #3

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7891. Min Act: -0.0001

Data Index: 497066 (C4 (Web Text))

Max Activating Token Index: 845

Text #4

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.8064. Min Act: -0.0001

Data Index: 191554 (C4 (Web Text))

Max Activating Token Index: 49

Text #5

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7833. Min Act: -0.0001

Data Index: 1600498 (Python Code)

Max Activating Token Index: 174

Text #6

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.8247. Min Act: -0.0001

Data Index: 520520 (C4 (Web Text))

Max Activating Token Index: 678

Text #7

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7681. Min Act: -0.0001

Data Index: 206709 (C4 (Web Text))

Max Activating Token Index: 971

Text #8

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7847. Min Act: -0.0001

Data Index: 583669 (C4 (Web Text))

Max Activating Token Index: 873

Text #9

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7863. Min Act: -0.0001

Data Index: 11453 (C4 (Web Text))

Max Activating Token Index: 730

Text #10

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7815. Min Act: -0.0001

Data Index: 1442746 (Python Code)

Max Activating Token Index: 117

Text #11

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.8048. Min Act: -0.0001

Data Index: 1361939 (Python Code)

Max Activating Token Index: 249

Text #12

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7567. Min Act: -0.0001

Data Index: 1223639 (C4 (Web Text))

Max Activating Token Index: 198

Text #13

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7988. Min Act: -0.0001

Data Index: 987436 (C4 (Web Text))

Max Activating Token Index: 301

Text #14

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7762. Min Act: -0.0001

Data Index: 587264 (C4 (Web Text))

Max Activating Token Index: 893

Text #15

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7597. Min Act: -0.0001

Data Index: 1388616 (Python Code)

Max Activating Token Index: 549

Text #16

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7804. Min Act: -0.0001

Data Index: 48962 (C4 (Web Text))

Max Activating Token Index: 995

Text #17

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.7684. Min Act: -0.0001

Data Index: 1115696 (C4 (Web Text))

Max Activating Token Index: 98

Text #18

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.6985. Min Act: -0.0001

Data Index: 810616 (C4 (Web Text))

Max Activating Token Index: 255

Text #19

Max Range: 0.8683. Min Range: -0.8683

Max Act: 0.6481. Min Act: -0.0001

Data Index: 450954 (C4 (Web Text))

Max Activating Token Index: 662