Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5699 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.7081. Min Act: -0.0000

Data Index: 274643 (C4 (Web Text))

Max Activating Token Index: 7

Text #1

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.7203. Min Act: -0.0000

Data Index: 886767 (C4 (Web Text))

Max Activating Token Index: 904

Text #2

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.7947. Min Act: -0.0000

Data Index: 1629503 (Python Code)

Max Activating Token Index: 375

Text #3

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.6253. Min Act: -0.0000

Data Index: 544247 (C4 (Web Text))

Max Activating Token Index: 265

Text #4

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.7537. Min Act: -0.0000

Data Index: 1568459 (Python Code)

Max Activating Token Index: 20

Text #5

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.5628. Min Act: -0.0000

Data Index: 59633 (C4 (Web Text))

Max Activating Token Index: 10

Text #6

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.6792. Min Act: -0.0000

Data Index: 647054 (C4 (Web Text))

Max Activating Token Index: 339

Text #7

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.6142. Min Act: -0.0000

Data Index: 1034700 (C4 (Web Text))

Max Activating Token Index: 171

Text #8

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.6399. Min Act: -0.0000

Data Index: 981294 (C4 (Web Text))

Max Activating Token Index: 275

Text #9

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.1570. Min Act: -0.0000

Data Index: 732489 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.4170. Min Act: -0.0000

Data Index: 1201356 (C4 (Web Text))

Max Activating Token Index: 354

Text #11

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.2808. Min Act: -0.0000

Data Index: 1011026 (C4 (Web Text))

Max Activating Token Index: 889

Text #12

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.4337. Min Act: -0.0000

Data Index: 488783 (C4 (Web Text))

Max Activating Token Index: 596

Text #13

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.4061. Min Act: -0.0000

Data Index: 1112730 (C4 (Web Text))

Max Activating Token Index: 73

Text #14

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.3705. Min Act: -0.0000

Data Index: 1651467 (Python Code)

Max Activating Token Index: 43

Text #15

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.3557. Min Act: -0.0000

Data Index: 725834 (C4 (Web Text))

Max Activating Token Index: 169

Text #16

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.3458. Min Act: -0.0000

Data Index: 1163276 (C4 (Web Text))

Max Activating Token Index: 235

Text #17

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.3394. Min Act: -0.0000

Data Index: 1493814 (Python Code)

Max Activating Token Index: 503

Text #18

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.4712. Min Act: -0.0000

Data Index: 1584238 (Python Code)

Max Activating Token Index: 841

Text #19

Max Range: 2.7947. Min Range: -2.7947

Max Act: 2.3042. Min Act: -0.0000

Data Index: 1453383 (Python Code)

Max Activating Token Index: 528