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

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

Neuron 4866 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: 0.6846. Min Range: -0.6846

Max Act: 0.6665. Min Act: -0.0001

Data Index: 353958 (C4 (Web Text))

Max Activating Token Index: 325

Text #1

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6846. Min Act: -0.0001

Data Index: 256730 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6387. Min Act: -0.0001

Data Index: 914760 (C4 (Web Text))

Max Activating Token Index: 566

Text #3

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6719. Min Act: -0.0001

Data Index: 501384 (C4 (Web Text))

Max Activating Token Index: 598

Text #4

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6546. Min Act: -0.0001

Data Index: 798139 (C4 (Web Text))

Max Activating Token Index: 127

Text #5

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6545. Min Act: -0.0001

Data Index: 1096861 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6086. Min Act: -0.0001

Data Index: 564658 (C4 (Web Text))

Max Activating Token Index: 675

Text #7

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6482. Min Act: -0.0001

Data Index: 969442 (C4 (Web Text))

Max Activating Token Index: 538

Text #8

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6206. Min Act: -0.0001

Data Index: 9721 (C4 (Web Text))

Max Activating Token Index: 368

Text #9

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6157. Min Act: -0.0001

Data Index: 1251952 (C4 (Web Text))

Max Activating Token Index: 370

Text #10

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.5961. Min Act: -0.0001

Data Index: 29635 (C4 (Web Text))

Max Activating Token Index: 529

Text #11

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6147. Min Act: -0.0001

Data Index: 770885 (C4 (Web Text))

Max Activating Token Index: 842

Text #12

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6477. Min Act: -0.0001

Data Index: 1016510 (C4 (Web Text))

Max Activating Token Index: 460

Text #13

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6018. Min Act: -0.0001

Data Index: 695574 (C4 (Web Text))

Max Activating Token Index: 627

Text #14

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.5943. Min Act: -0.0001

Data Index: 193092 (C4 (Web Text))

Max Activating Token Index: 400

Text #15

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6210. Min Act: -0.0001

Data Index: 1487947 (Python Code)

Max Activating Token Index: 524

Text #16

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.5464. Min Act: -0.0001

Data Index: 1366402 (Python Code)

Max Activating Token Index: 212

Text #17

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6141. Min Act: -0.0001

Data Index: 52742 (C4 (Web Text))

Max Activating Token Index: 573

Text #18

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.5881. Min Act: -0.0001

Data Index: 1245326 (C4 (Web Text))

Max Activating Token Index: 579

Text #19

Max Range: 0.6846. Min Range: -0.6846

Max Act: 0.6306. Min Act: -0.0001

Data Index: 606792 (C4 (Web Text))

Max Activating Token Index: 185