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

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

Neuron 1273 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.8168. Min Act: -0.0002

Data Index: 923255 (C4 (Web Text))

Max Activating Token Index: 57

Text #1

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.8430. Min Act: -0.0002

Data Index: 106394 (C4 (Web Text))

Max Activating Token Index: 295

Text #2

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.8528. Min Act: -0.0001

Data Index: 534288 (C4 (Web Text))

Max Activating Token Index: 513

Text #3

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.8219. Min Act: -0.0002

Data Index: 947617 (C4 (Web Text))

Max Activating Token Index: 150

Text #4

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7041. Min Act: -0.0002

Data Index: 554584 (C4 (Web Text))

Max Activating Token Index: 610

Text #5

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7412. Min Act: -0.0002

Data Index: 827972 (C4 (Web Text))

Max Activating Token Index: 488

Text #6

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7141. Min Act: -0.0002

Data Index: 698057 (C4 (Web Text))

Max Activating Token Index: 22

Text #7

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6465. Min Act: -0.0002

Data Index: 298027 (C4 (Web Text))

Max Activating Token Index: 130

Text #8

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7515. Min Act: -0.0001

Data Index: 140211 (C4 (Web Text))

Max Activating Token Index: 596

Text #9

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7639. Min Act: -0.0002

Data Index: 34199 (C4 (Web Text))

Max Activating Token Index: 856

Text #10

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6844. Min Act: -0.0002

Data Index: 883725 (C4 (Web Text))

Max Activating Token Index: 388

Text #11

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7269. Min Act: -0.0002

Data Index: 609010 (C4 (Web Text))

Max Activating Token Index: 99

Text #12

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7244. Min Act: -0.0002

Data Index: 169392 (C4 (Web Text))

Max Activating Token Index: 590

Text #13

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7606. Min Act: -0.0002

Data Index: 205918 (C4 (Web Text))

Max Activating Token Index: 723

Text #14

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7172. Min Act: -0.0002

Data Index: 233308 (C4 (Web Text))

Max Activating Token Index: 383

Text #15

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.7114. Min Act: -0.0002

Data Index: 461374 (C4 (Web Text))

Max Activating Token Index: 764

Text #16

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6850. Min Act: -0.0001

Data Index: 858129 (C4 (Web Text))

Max Activating Token Index: 783

Text #17

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6934. Min Act: -0.0002

Data Index: 806048 (C4 (Web Text))

Max Activating Token Index: 759

Text #18

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6819. Min Act: -0.0001

Data Index: 484157 (C4 (Web Text))

Max Activating Token Index: 301

Text #19

Max Range: 0.8528. Min Range: -0.8528

Max Act: 0.6832. Min Act: -0.0002

Data Index: 1054521 (C4 (Web Text))

Max Activating Token Index: 88