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

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

Neuron 3077 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: 1.4020. Min Range: -1.4020

Max Act: 1.4020. Min Act: -0.0001

Data Index: 585069 (C4 (Web Text))

Max Activating Token Index: 254

Text #1

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.3213. Min Act: -0.0001

Data Index: 1476089 (Python Code)

Max Activating Token Index: 279

Text #2

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.3597. Min Act: -0.0001

Data Index: 424890 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.1675. Min Act: -0.0001

Data Index: 1048254 (C4 (Web Text))

Max Activating Token Index: 695

Text #4

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.1536. Min Act: -0.0001

Data Index: 1384157 (Python Code)

Max Activating Token Index: 707

Text #5

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0632. Min Act: -0.0001

Data Index: 133592 (C4 (Web Text))

Max Activating Token Index: 231

Text #6

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.2389. Min Act: -0.0001

Data Index: 746399 (C4 (Web Text))

Max Activating Token Index: 811

Text #7

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.2136. Min Act: -0.0001

Data Index: 1326693 (C4 (Web Text))

Max Activating Token Index: 64

Text #8

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.2093. Min Act: -0.0001

Data Index: 898651 (C4 (Web Text))

Max Activating Token Index: 728

Text #9

Max Range: 1.4020. Min Range: -1.4020

Max Act: 0.9974. Min Act: -0.0001

Data Index: 50425 (C4 (Web Text))

Max Activating Token Index: 69

Text #10

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.1579. Min Act: -0.0001

Data Index: 1539471 (Python Code)

Max Activating Token Index: 175

Text #11

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0904. Min Act: -0.0001

Data Index: 529635 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0443. Min Act: -0.0001

Data Index: 976884 (C4 (Web Text))

Max Activating Token Index: 237

Text #13

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0672. Min Act: -0.0001

Data Index: 575111 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0921. Min Act: -0.0001

Data Index: 1301381 (C4 (Web Text))

Max Activating Token Index: 714

Text #15

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0739. Min Act: -0.0001

Data Index: 102637 (C4 (Web Text))

Max Activating Token Index: 631

Text #16

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.1222. Min Act: -0.0001

Data Index: 572045 (C4 (Web Text))

Max Activating Token Index: 650

Text #17

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0472. Min Act: -0.0001

Data Index: 1492769 (Python Code)

Max Activating Token Index: 829

Text #18

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0984. Min Act: -0.0001

Data Index: 125673 (C4 (Web Text))

Max Activating Token Index: 616

Text #19

Max Range: 1.4020. Min Range: -1.4020

Max Act: 1.0460. Min Act: -0.0001

Data Index: 585936 (C4 (Web Text))

Max Activating Token Index: 451