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

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

Neuron 681 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.4609. Min Act: -0.0001

Data Index: 1063934 (C4 (Web Text))

Max Activating Token Index: 208

Text #1

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.4279. Min Act: -0.0001

Data Index: 319468 (C4 (Web Text))

Max Activating Token Index: 952

Text #2

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.3883. Min Act: -0.0001

Data Index: 1106330 (C4 (Web Text))

Max Activating Token Index: 567

Text #3

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2885. Min Act: -0.0001

Data Index: 1198619 (C4 (Web Text))

Max Activating Token Index: 334

Text #4

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2668. Min Act: -0.0001

Data Index: 1308773 (C4 (Web Text))

Max Activating Token Index: 224

Text #5

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2838. Min Act: -0.0001

Data Index: 360099 (C4 (Web Text))

Max Activating Token Index: 120

Text #6

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2954. Min Act: -0.0001

Data Index: 439101 (C4 (Web Text))

Max Activating Token Index: 158

Text #7

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2935. Min Act: -0.0001

Data Index: 965304 (C4 (Web Text))

Max Activating Token Index: 168

Text #8

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.3066. Min Act: -0.0001

Data Index: 552864 (C4 (Web Text))

Max Activating Token Index: 112

Text #9

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.3144. Min Act: -0.0001

Data Index: 922214 (C4 (Web Text))

Max Activating Token Index: 98

Text #10

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2913. Min Act: -0.0001

Data Index: 225149 (C4 (Web Text))

Max Activating Token Index: 331

Text #11

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2775. Min Act: -0.0001

Data Index: 1510927 (Python Code)

Max Activating Token Index: 61

Text #12

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2679. Min Act: -0.0001

Data Index: 1244015 (C4 (Web Text))

Max Activating Token Index: 336

Text #13

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2304. Min Act: -0.0001

Data Index: 739127 (C4 (Web Text))

Max Activating Token Index: 127

Text #14

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2754. Min Act: -0.0001

Data Index: 562688 (C4 (Web Text))

Max Activating Token Index: 8

Text #15

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2986. Min Act: -0.0001

Data Index: 1090347 (C4 (Web Text))

Max Activating Token Index: 236

Text #16

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2931. Min Act: -0.0001

Data Index: 825057 (C4 (Web Text))

Max Activating Token Index: 140

Text #17

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2375. Min Act: -0.0001

Data Index: 760191 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2642. Min Act: -0.0001

Data Index: 419011 (C4 (Web Text))

Max Activating Token Index: 177

Text #19

Max Range: 0.4609. Min Range: -0.4609

Max Act: 0.2538. Min Act: -0.0001

Data Index: 478656 (C4 (Web Text))

Max Activating Token Index: 11