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

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

Neuron 4439 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.1937. Min Range: -6.1937

Max Act: 6.1937. Min Act: -0.0001

Data Index: 1270955 (C4 (Web Text))

Max Activating Token Index: 357

Text #1

Max Range: 6.1937. Min Range: -6.1937

Max Act: 6.0029. Min Act: -0.0001

Data Index: 724766 (C4 (Web Text))

Max Activating Token Index: 361

Text #2

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.8210. Min Act: -0.0001

Data Index: 970148 (C4 (Web Text))

Max Activating Token Index: 570

Text #3

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.9496. Min Act: -0.0001

Data Index: 218438 (C4 (Web Text))

Max Activating Token Index: 669

Text #4

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.5966. Min Act: -0.0001

Data Index: 1308698 (C4 (Web Text))

Max Activating Token Index: 896

Text #5

Max Range: 6.1937. Min Range: -6.1937

Max Act: 6.1385. Min Act: -0.0001

Data Index: 677932 (C4 (Web Text))

Max Activating Token Index: 571

Text #6

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.9480. Min Act: -0.0001

Data Index: 1348459 (C4 (Web Text))

Max Activating Token Index: 200

Text #7

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.6628. Min Act: -0.0001

Data Index: 304127 (C4 (Web Text))

Max Activating Token Index: 431

Text #8

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.6788. Min Act: -0.0001

Data Index: 17409 (C4 (Web Text))

Max Activating Token Index: 646

Text #9

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.8441. Min Act: -0.0001

Data Index: 529013 (C4 (Web Text))

Max Activating Token Index: 332

Text #10

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.8119. Min Act: -0.0001

Data Index: 256180 (C4 (Web Text))

Max Activating Token Index: 728

Text #11

Max Range: 6.1937. Min Range: -6.1937

Max Act: 5.5615. Min Act: -0.0001

Data Index: 470447 (C4 (Web Text))

Max Activating Token Index: 326

Text #12

Max Range: 6.1937. Min Range: -6.1937

Max Act: 4.8966. Min Act: -0.0001

Data Index: 305656 (C4 (Web Text))

Max Activating Token Index: 265

Text #13

Max Range: 6.1937. Min Range: -6.1937

Max Act: 4.1858. Min Act: -0.0001

Data Index: 1186745 (C4 (Web Text))

Max Activating Token Index: 321

Text #14

Max Range: 6.1937. Min Range: -6.1937

Max Act: 4.1913. Min Act: -0.0001

Data Index: 1143080 (C4 (Web Text))

Max Activating Token Index: 971

Text #15

Max Range: 6.1937. Min Range: -6.1937

Max Act: 3.9422. Min Act: -0.0001

Data Index: 224750 (C4 (Web Text))

Max Activating Token Index: 26

Text #16

Max Range: 6.1937. Min Range: -6.1937

Max Act: 3.8274. Min Act: -0.0001

Data Index: 931789 (C4 (Web Text))

Max Activating Token Index: 320

Text #17

Max Range: 6.1937. Min Range: -6.1937

Max Act: 3.6603. Min Act: -0.0001

Data Index: 1031802 (C4 (Web Text))

Max Activating Token Index: 9

Text #18

Max Range: 6.1937. Min Range: -6.1937

Max Act: 3.8296. Min Act: -0.0001

Data Index: 1116343 (C4 (Web Text))

Max Activating Token Index: 163

Text #19

Max Range: 6.1937. Min Range: -6.1937

Max Act: 3.3116. Min Act: -0.0001

Data Index: 823819 (C4 (Web Text))

Max Activating Token Index: 956