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

Dataset: The Pile

Neuron 1364 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.4129. Min Act: -0.0000

Data Index: 1439231 (The Pile)

Max Activating Token Index: 187

Text #1

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.4190. Min Act: -0.0000

Data Index: 374304 (The Pile)

Max Activating Token Index: 449

Text #2

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.7013. Min Act: -0.0000

Data Index: 450478 (The Pile)

Max Activating Token Index: 755

Text #3

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.9979. Min Act: -0.0000

Data Index: 799715 (The Pile)

Max Activating Token Index: 85

Text #4

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.1398. Min Act: -0.0000

Data Index: 674803 (The Pile)

Max Activating Token Index: 772

Text #5

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.1107. Min Act: -0.0000

Data Index: 501683 (The Pile)

Max Activating Token Index: 211

Text #6

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.6419. Min Act: -0.0000

Data Index: 768509 (The Pile)

Max Activating Token Index: 123

Text #7

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.9576. Min Act: -0.0000

Data Index: 1348592 (The Pile)

Max Activating Token Index: 573

Text #8

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.5934. Min Act: -0.0000

Data Index: 95380 (The Pile)

Max Activating Token Index: 563

Text #9

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.3402. Min Act: -0.0000

Data Index: 1978406 (The Pile)

Max Activating Token Index: 875

Text #10

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.3171. Min Act: -0.0000

Data Index: 349922 (The Pile)

Max Activating Token Index: 126

Text #11

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.7786. Min Act: -0.0000

Data Index: 1813045 (The Pile)

Max Activating Token Index: 82

Text #12

Max Range: 9.4190. Min Range: -9.4190

Max Act: 9.1966. Min Act: -0.0000

Data Index: 1953624 (The Pile)

Max Activating Token Index: 404

Text #13

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.5063. Min Act: -0.0000

Data Index: 663221 (The Pile)

Max Activating Token Index: 134

Text #14

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.7635. Min Act: -0.0000

Data Index: 409755 (The Pile)

Max Activating Token Index: 83

Text #15

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.4027. Min Act: -0.0000

Data Index: 156147 (The Pile)

Max Activating Token Index: 343

Text #16

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.5287. Min Act: -0.0000

Data Index: 1737653 (The Pile)

Max Activating Token Index: 203

Text #17

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.4812. Min Act: -0.0000

Data Index: 1294150 (The Pile)

Max Activating Token Index: 671

Text #18

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.2798. Min Act: -0.0000

Data Index: 1226268 (The Pile)

Max Activating Token Index: 838

Text #19

Max Range: 9.4190. Min Range: -9.4190

Max Act: 8.4458. Min Act: -0.0000

Data Index: 447267 (The Pile)

Max Activating Token Index: 587