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

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

Neuron 1478 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0295. Min Act: -0.0001

Data Index: 1383723 (Python Code)

Max Activating Token Index: 453

Text #1

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0209. Min Act: -0.0001

Data Index: 291518 (C4 (Web Text))

Max Activating Token Index: 16

Text #2

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0214. Min Act: -0.0001

Data Index: 1626251 (Python Code)

Max Activating Token Index: 973

Text #3

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0187. Min Act: -0.0001

Data Index: 887397 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0193. Min Act: -0.0001

Data Index: 533016 (C4 (Web Text))

Max Activating Token Index: 433

Text #5

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0204. Min Act: -0.0001

Data Index: 190592 (C4 (Web Text))

Max Activating Token Index: 1006

Text #6

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0197. Min Act: -0.0001

Data Index: 1331772 (C4 (Web Text))

Max Activating Token Index: 149

Text #7

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0160. Min Act: -0.0001

Data Index: 1392506 (Python Code)

Max Activating Token Index: 413

Text #8

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0184. Min Act: -0.0001

Data Index: 194569 (C4 (Web Text))

Max Activating Token Index: 707

Text #9

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0186. Min Act: -0.0001

Data Index: 983544 (C4 (Web Text))

Max Activating Token Index: 346

Text #10

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0190. Min Act: -0.0000

Data Index: 473686 (C4 (Web Text))

Max Activating Token Index: 879

Text #11

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0196. Min Act: -0.0001

Data Index: 75282 (C4 (Web Text))

Max Activating Token Index: 929

Text #12

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0183. Min Act: -0.0001

Data Index: 1012736 (C4 (Web Text))

Max Activating Token Index: 587

Text #13

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0177. Min Act: -0.0001

Data Index: 883299 (C4 (Web Text))

Max Activating Token Index: 187

Text #14

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0175. Min Act: -0.0001

Data Index: 198220 (C4 (Web Text))

Max Activating Token Index: 498

Text #15

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0169. Min Act: -0.0001

Data Index: 586381 (C4 (Web Text))

Max Activating Token Index: 260

Text #16

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0182. Min Act: -0.0001

Data Index: 1540558 (Python Code)

Max Activating Token Index: 456

Text #17

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0190. Min Act: -0.0001

Data Index: 1040933 (C4 (Web Text))

Max Activating Token Index: 537

Text #18

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0188. Min Act: -0.0001

Data Index: 499402 (C4 (Web Text))

Max Activating Token Index: 9

Text #19

Max Range: 0.0295. Min Range: -0.0295

Max Act: 0.0169. Min Act: -0.0001

Data Index: 465638 (C4 (Web Text))

Max Activating Token Index: 203