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

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

Neuron 3935 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: 3.7188. Min Range: -3.7188

Max Act: 3.7188. Min Act: -0.0001

Data Index: 838563 (C4 (Web Text))

Max Activating Token Index: 821

Text #1

Max Range: 3.7188. Min Range: -3.7188

Max Act: 3.6123. Min Act: -0.0001

Data Index: 200333 (C4 (Web Text))

Max Activating Token Index: 17

Text #2

Max Range: 3.7188. Min Range: -3.7188

Max Act: 3.2170. Min Act: -0.0001

Data Index: 27771 (C4 (Web Text))

Max Activating Token Index: 532

Text #3

Max Range: 3.7188. Min Range: -3.7188

Max Act: 3.0128. Min Act: -0.0001

Data Index: 594042 (C4 (Web Text))

Max Activating Token Index: 918

Text #4

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.9741. Min Act: -0.0001

Data Index: 16224 (C4 (Web Text))

Max Activating Token Index: 978

Text #5

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8012. Min Act: -0.0001

Data Index: 485647 (C4 (Web Text))

Max Activating Token Index: 525

Text #6

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.7810. Min Act: -0.0001

Data Index: 156541 (C4 (Web Text))

Max Activating Token Index: 262

Text #7

Max Range: 3.7188. Min Range: -3.7188

Max Act: 3.1989. Min Act: -0.0001

Data Index: 647074 (C4 (Web Text))

Max Activating Token Index: 617

Text #8

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.9159. Min Act: -0.0001

Data Index: 35798 (C4 (Web Text))

Max Activating Token Index: 790

Text #9

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8564. Min Act: -0.0001

Data Index: 576262 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8257. Min Act: -0.0001

Data Index: 543576 (C4 (Web Text))

Max Activating Token Index: 120

Text #11

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8175. Min Act: -0.0001

Data Index: 1048544 (C4 (Web Text))

Max Activating Token Index: 717

Text #12

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8273. Min Act: -0.0001

Data Index: 881919 (C4 (Web Text))

Max Activating Token Index: 339

Text #13

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.6278. Min Act: -0.0001

Data Index: 650194 (C4 (Web Text))

Max Activating Token Index: 864

Text #14

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.8637. Min Act: -0.0001

Data Index: 618655 (C4 (Web Text))

Max Activating Token Index: 935

Text #15

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.5752. Min Act: -0.0001

Data Index: 331212 (C4 (Web Text))

Max Activating Token Index: 901

Text #16

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.7173. Min Act: -0.0001

Data Index: 813037 (C4 (Web Text))

Max Activating Token Index: 142

Text #17

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.5478. Min Act: -0.0001

Data Index: 947640 (C4 (Web Text))

Max Activating Token Index: 849

Text #18

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.6564. Min Act: -0.0001

Data Index: 776533 (C4 (Web Text))

Max Activating Token Index: 499

Text #19

Max Range: 3.7188. Min Range: -3.7188

Max Act: 2.6408. Min Act: -0.0001

Data Index: 758785 (C4 (Web Text))

Max Activating Token Index: 801