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

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

Neuron 3501 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2487. Min Act: -0.0001

Data Index: 1459469 (Python Code)

Max Activating Token Index: 66

Text #1

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2485. Min Act: -0.0001

Data Index: 1142026 (C4 (Web Text))

Max Activating Token Index: 882

Text #2

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2381. Min Act: -0.0001

Data Index: 1167691 (C4 (Web Text))

Max Activating Token Index: 481

Text #3

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2315. Min Act: -0.0001

Data Index: 1073557 (C4 (Web Text))

Max Activating Token Index: 1000

Text #4

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2318. Min Act: -0.0001

Data Index: 1547074 (Python Code)

Max Activating Token Index: 228

Text #5

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2277. Min Act: -0.0001

Data Index: 1164387 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2241. Min Act: -0.0001

Data Index: 685068 (C4 (Web Text))

Max Activating Token Index: 187

Text #7

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2290. Min Act: -0.0001

Data Index: 1322290 (C4 (Web Text))

Max Activating Token Index: 470

Text #8

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2195. Min Act: -0.0001

Data Index: 975104 (C4 (Web Text))

Max Activating Token Index: 376

Text #9

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2182. Min Act: -0.0001

Data Index: 739057 (C4 (Web Text))

Max Activating Token Index: 760

Text #10

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2104. Min Act: -0.0001

Data Index: 636551 (C4 (Web Text))

Max Activating Token Index: 225

Text #11

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2333. Min Act: -0.0001

Data Index: 1427482 (Python Code)

Max Activating Token Index: 665

Text #12

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2146. Min Act: -0.0001

Data Index: 106387 (C4 (Web Text))

Max Activating Token Index: 346

Text #13

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2069. Min Act: -0.0001

Data Index: 432947 (C4 (Web Text))

Max Activating Token Index: 726

Text #14

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2182. Min Act: -0.0001

Data Index: 1315935 (C4 (Web Text))

Max Activating Token Index: 782

Text #15

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2022. Min Act: -0.0001

Data Index: 1558081 (Python Code)

Max Activating Token Index: 369

Text #16

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2097. Min Act: -0.0001

Data Index: 1531571 (Python Code)

Max Activating Token Index: 814

Text #17

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2009. Min Act: -0.0001

Data Index: 422463 (C4 (Web Text))

Max Activating Token Index: 834

Text #18

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2094. Min Act: -0.0001

Data Index: 241252 (C4 (Web Text))

Max Activating Token Index: 398

Text #19

Max Range: 0.2487. Min Range: -0.2487

Max Act: 0.2075. Min Act: -0.0001

Data Index: 1455208 (Python Code)

Max Activating Token Index: 909