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

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

Neuron 4999 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3441. Min Range: -1.3441

Max Act: 1.3403. Min Act: -0.0000

Data Index: 1574552 (Python Code)

Max Activating Token Index: 565

Text #1

Max Range: 1.3441. Min Range: -1.3441

Max Act: 1.3441. Min Act: -0.0000

Data Index: 26645 (C4 (Web Text))

Max Activating Token Index: 832

Text #2

Max Range: 1.3441. Min Range: -1.3441

Max Act: 1.1031. Min Act: -0.0000

Data Index: 1173356 (C4 (Web Text))

Max Activating Token Index: 74

Text #3

Max Range: 1.3441. Min Range: -1.3441

Max Act: 1.0166. Min Act: -0.0000

Data Index: 410705 (C4 (Web Text))

Max Activating Token Index: 269

Text #4

Max Range: 1.3441. Min Range: -1.3441

Max Act: 1.0241. Min Act: -0.0000

Data Index: 503387 (C4 (Web Text))

Max Activating Token Index: 101

Text #5

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.9721. Min Act: -0.0000

Data Index: 1098438 (C4 (Web Text))

Max Activating Token Index: 470

Text #6

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.9585. Min Act: -0.0000

Data Index: 852637 (C4 (Web Text))

Max Activating Token Index: 357

Text #7

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.9308. Min Act: -0.0000

Data Index: 918366 (C4 (Web Text))

Max Activating Token Index: 705

Text #8

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.9042. Min Act: -0.0000

Data Index: 555042 (C4 (Web Text))

Max Activating Token Index: 349

Text #9

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8939. Min Act: -0.0000

Data Index: 855433 (C4 (Web Text))

Max Activating Token Index: 455

Text #10

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.9202. Min Act: -0.0000

Data Index: 183265 (C4 (Web Text))

Max Activating Token Index: 213

Text #11

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8888. Min Act: -0.0000

Data Index: 1251837 (C4 (Web Text))

Max Activating Token Index: 614

Text #12

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8284. Min Act: -0.0000

Data Index: 1432311 (Python Code)

Max Activating Token Index: 982

Text #13

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8988. Min Act: -0.0000

Data Index: 654928 (C4 (Web Text))

Max Activating Token Index: 1003

Text #14

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8082. Min Act: -0.0000

Data Index: 1249453 (C4 (Web Text))

Max Activating Token Index: 843

Text #15

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.7655. Min Act: -0.0000

Data Index: 1394582 (Python Code)

Max Activating Token Index: 904

Text #16

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8768. Min Act: -0.0000

Data Index: 461007 (C4 (Web Text))

Max Activating Token Index: 330

Text #17

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8090. Min Act: -0.0000

Data Index: 838990 (C4 (Web Text))

Max Activating Token Index: 43

Text #18

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8007. Min Act: -0.0000

Data Index: 992221 (C4 (Web Text))

Max Activating Token Index: 792

Text #19

Max Range: 1.3441. Min Range: -1.3441

Max Act: 0.8149. Min Act: -0.0000

Data Index: 1587954 (Python Code)

Max Activating Token Index: 324