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

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

Neuron 3621 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2170. Min Range: -1.2170

Max Act: 1.0334. Min Act: -0.0001

Data Index: 394204 (C4 (Web Text))

Max Activating Token Index: 9

Text #1

Max Range: 1.2170. Min Range: -1.2170

Max Act: 1.2170. Min Act: -0.0001

Data Index: 625323 (C4 (Web Text))

Max Activating Token Index: 341

Text #2

Max Range: 1.2170. Min Range: -1.2170

Max Act: 1.0013. Min Act: -0.0001

Data Index: 1071160 (C4 (Web Text))

Max Activating Token Index: 213

Text #3

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.9213. Min Act: -0.0001

Data Index: 717020 (C4 (Web Text))

Max Activating Token Index: 473

Text #4

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.9411. Min Act: -0.0001

Data Index: 1083230 (C4 (Web Text))

Max Activating Token Index: 222

Text #5

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.9339. Min Act: -0.0001

Data Index: 479386 (C4 (Web Text))

Max Activating Token Index: 645

Text #6

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.6253. Min Act: -0.0001

Data Index: 1020596 (C4 (Web Text))

Max Activating Token Index: 260

Text #7

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.8734. Min Act: -0.0001

Data Index: 21811 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7145. Min Act: -0.0001

Data Index: 1645399 (Python Code)

Max Activating Token Index: 316

Text #9

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.8285. Min Act: -0.0001

Data Index: 1400790 (Python Code)

Max Activating Token Index: 684

Text #10

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7945. Min Act: -0.0001

Data Index: 841208 (C4 (Web Text))

Max Activating Token Index: 438

Text #11

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7491. Min Act: -0.0001

Data Index: 930173 (C4 (Web Text))

Max Activating Token Index: 120

Text #12

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7932. Min Act: -0.0001

Data Index: 1131632 (C4 (Web Text))

Max Activating Token Index: 409

Text #13

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7618. Min Act: -0.0000

Data Index: 452675 (C4 (Web Text))

Max Activating Token Index: 864

Text #14

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.8388. Min Act: -0.0001

Data Index: 1558455 (Python Code)

Max Activating Token Index: 187

Text #15

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7277. Min Act: -0.0001

Data Index: 1584238 (Python Code)

Max Activating Token Index: 905

Text #16

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7817. Min Act: -0.0001

Data Index: 117397 (C4 (Web Text))

Max Activating Token Index: 432

Text #17

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7989. Min Act: -0.0001

Data Index: 1531768 (Python Code)

Max Activating Token Index: 1018

Text #18

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7437. Min Act: -0.0001

Data Index: 566769 (C4 (Web Text))

Max Activating Token Index: 741

Text #19

Max Range: 1.2170. Min Range: -1.2170

Max Act: 0.7398. Min Act: -0.0001

Data Index: 361579 (C4 (Web Text))

Max Activating Token Index: 484