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

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

Neuron 3463 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: 1.0226. Min Range: -1.0226

Max Act: 0.9713. Min Act: -0.0001

Data Index: 929171 (C4 (Web Text))

Max Activating Token Index: 840

Text #1

Max Range: 1.0226. Min Range: -1.0226

Max Act: 1.0226. Min Act: -0.0001

Data Index: 1577811 (Python Code)

Max Activating Token Index: 4

Text #2

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.9492. Min Act: -0.0001

Data Index: 885466 (C4 (Web Text))

Max Activating Token Index: 146

Text #3

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.9610. Min Act: -0.0001

Data Index: 604309 (C4 (Web Text))

Max Activating Token Index: 888

Text #4

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8767. Min Act: -0.0001

Data Index: 1341645 (C4 (Web Text))

Max Activating Token Index: 707

Text #5

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8870. Min Act: -0.0001

Data Index: 875154 (C4 (Web Text))

Max Activating Token Index: 5

Text #6

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8584. Min Act: -0.0001

Data Index: 521461 (C4 (Web Text))

Max Activating Token Index: 832

Text #7

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.9030. Min Act: -0.0001

Data Index: 1091238 (C4 (Web Text))

Max Activating Token Index: 565

Text #8

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8119. Min Act: -0.0001

Data Index: 482756 (C4 (Web Text))

Max Activating Token Index: 917

Text #9

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8713. Min Act: -0.0001

Data Index: 51433 (C4 (Web Text))

Max Activating Token Index: 674

Text #10

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7906. Min Act: -0.0001

Data Index: 299933 (C4 (Web Text))

Max Activating Token Index: 529

Text #11

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8666. Min Act: -0.0001

Data Index: 837837 (C4 (Web Text))

Max Activating Token Index: 285

Text #12

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8819. Min Act: -0.0001

Data Index: 644329 (C4 (Web Text))

Max Activating Token Index: 707

Text #13

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7825. Min Act: -0.0001

Data Index: 1331017 (C4 (Web Text))

Max Activating Token Index: 368

Text #14

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8139. Min Act: -0.0001

Data Index: 901927 (C4 (Web Text))

Max Activating Token Index: 636

Text #15

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.8172. Min Act: -0.0001

Data Index: 218503 (C4 (Web Text))

Max Activating Token Index: 968

Text #16

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7898. Min Act: -0.0001

Data Index: 209860 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7850. Min Act: -0.0001

Data Index: 810883 (C4 (Web Text))

Max Activating Token Index: 286

Text #18

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7788. Min Act: -0.0001

Data Index: 463880 (C4 (Web Text))

Max Activating Token Index: 193

Text #19

Max Range: 1.0226. Min Range: -1.0226

Max Act: 0.7761. Min Act: -0.0001

Data Index: 1149113 (C4 (Web Text))

Max Activating Token Index: 35