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

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

Neuron 3948 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0120. Min Act: -0.0004

Data Index: 331313 (C4 (Web Text))

Max Activating Token Index: 65

Text #1

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0105. Min Act: -0.0004

Data Index: 494677 (C4 (Web Text))

Max Activating Token Index: 306

Text #2

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0101. Min Act: -0.0004

Data Index: 354073 (C4 (Web Text))

Max Activating Token Index: 103

Text #3

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0111. Min Act: -0.0003

Data Index: 789448 (C4 (Web Text))

Max Activating Token Index: 318

Text #4

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0101. Min Act: -0.0004

Data Index: 619867 (C4 (Web Text))

Max Activating Token Index: 96

Text #5

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0093. Min Act: -0.0003

Data Index: 1108529 (C4 (Web Text))

Max Activating Token Index: 847

Text #6

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0093. Min Act: -0.0003

Data Index: 1040498 (C4 (Web Text))

Max Activating Token Index: 298

Text #7

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0110. Min Act: -0.0004

Data Index: 353234 (C4 (Web Text))

Max Activating Token Index: 498

Text #8

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0100. Min Act: -0.0004

Data Index: 763065 (C4 (Web Text))

Max Activating Token Index: 262

Text #9

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0093. Min Act: -0.0004

Data Index: 790176 (C4 (Web Text))

Max Activating Token Index: 870

Text #10

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0093. Min Act: -0.0004

Data Index: 928726 (C4 (Web Text))

Max Activating Token Index: 121

Text #11

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0092. Min Act: -0.0003

Data Index: 1261025 (C4 (Web Text))

Max Activating Token Index: 415

Text #12

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0092. Min Act: -0.0003

Data Index: 345309 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0090. Min Act: -0.0004

Data Index: 530578 (C4 (Web Text))

Max Activating Token Index: 120

Text #14

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0093. Min Act: -0.0003

Data Index: 1648753 (Python Code)

Max Activating Token Index: 2

Text #15

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0094. Min Act: -0.0003

Data Index: 373758 (C4 (Web Text))

Max Activating Token Index: 388

Text #16

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0097. Min Act: -0.0003

Data Index: 1100179 (C4 (Web Text))

Max Activating Token Index: 35

Text #17

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0094. Min Act: -0.0003

Data Index: 235290 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0087. Min Act: -0.0004

Data Index: 703837 (C4 (Web Text))

Max Activating Token Index: 988

Text #19

Max Range: 0.0120. Min Range: -0.0120

Max Act: 0.0077. Min Act: -0.0004

Data Index: 109907 (C4 (Web Text))

Max Activating Token Index: 86