Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 520 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.3545. Min Act: -0.0001

Data Index: 195028 (C4 (Web Text))

Max Activating Token Index: 552

Text #1

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4625. Min Act: -0.0001

Data Index: 446437 (C4 (Web Text))

Max Activating Token Index: 459

Text #2

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4616. Min Act: -0.0001

Data Index: 655267 (C4 (Web Text))

Max Activating Token Index: 356

Text #3

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4526. Min Act: -0.0001

Data Index: 461570 (C4 (Web Text))

Max Activating Token Index: 346

Text #4

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4382. Min Act: -0.0001

Data Index: 537055 (C4 (Web Text))

Max Activating Token Index: 15

Text #5

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4704. Min Act: -0.0001

Data Index: 785228 (C4 (Web Text))

Max Activating Token Index: 21

Text #6

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.3945. Min Act: -0.0001

Data Index: 1308305 (C4 (Web Text))

Max Activating Token Index: 691

Text #7

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4304. Min Act: -0.0001

Data Index: 60258 (C4 (Web Text))

Max Activating Token Index: 312

Text #8

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4167. Min Act: -0.0001

Data Index: 1203128 (C4 (Web Text))

Max Activating Token Index: 727

Text #9

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4202. Min Act: -0.0001

Data Index: 214036 (C4 (Web Text))

Max Activating Token Index: 952

Text #10

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4035. Min Act: -0.0001

Data Index: 84307 (C4 (Web Text))

Max Activating Token Index: 83

Text #11

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4268. Min Act: -0.0001

Data Index: 449490 (C4 (Web Text))

Max Activating Token Index: 751

Text #12

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4130. Min Act: -0.0001

Data Index: 548449 (C4 (Web Text))

Max Activating Token Index: 61

Text #13

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4214. Min Act: -0.0001

Data Index: 392754 (C4 (Web Text))

Max Activating Token Index: 70

Text #14

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.3951. Min Act: -0.0001

Data Index: 728414 (C4 (Web Text))

Max Activating Token Index: 137

Text #15

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4218. Min Act: -0.0001

Data Index: 566569 (C4 (Web Text))

Max Activating Token Index: 853

Text #16

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4072. Min Act: -0.0001

Data Index: 1036213 (C4 (Web Text))

Max Activating Token Index: 235

Text #17

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.3934. Min Act: -0.0001

Data Index: 820917 (C4 (Web Text))

Max Activating Token Index: 419

Text #18

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.3885. Min Act: -0.0001

Data Index: 1183109 (C4 (Web Text))

Max Activating Token Index: 1019

Text #19

Max Range: 0.4704. Min Range: -0.4704

Max Act: 0.4059. Min Act: -0.0001

Data Index: 948344 (C4 (Web Text))

Max Activating Token Index: 636