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

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

Neuron 3634 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.8209. Min Act: -0.0000

Data Index: 693272 (C4 (Web Text))

Max Activating Token Index: 772

Text #1

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.7493. Min Act: -0.0000

Data Index: 95104 (C4 (Web Text))

Max Activating Token Index: 917

Text #2

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.7131. Min Act: -0.0000

Data Index: 777704 (C4 (Web Text))

Max Activating Token Index: 858

Text #3

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.6073. Min Act: -0.0000

Data Index: 1403401 (Python Code)

Max Activating Token Index: 300

Text #4

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.6461. Min Act: -0.0000

Data Index: 1359608 (C4 (Web Text))

Max Activating Token Index: 461

Text #5

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.7431. Min Act: -0.0000

Data Index: 1218232 (C4 (Web Text))

Max Activating Token Index: 736

Text #6

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.6898. Min Act: -0.0000

Data Index: 733379 (C4 (Web Text))

Max Activating Token Index: 749

Text #7

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.6831. Min Act: -0.0000

Data Index: 666955 (C4 (Web Text))

Max Activating Token Index: 159

Text #8

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5726. Min Act: -0.0000

Data Index: 677768 (C4 (Web Text))

Max Activating Token Index: 325

Text #9

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.6505. Min Act: -0.0000

Data Index: 1106283 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5918. Min Act: -0.0000

Data Index: 531528 (C4 (Web Text))

Max Activating Token Index: 933

Text #11

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5731. Min Act: -0.0000

Data Index: 465835 (C4 (Web Text))

Max Activating Token Index: 67

Text #12

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5936. Min Act: -0.0000

Data Index: 1620665 (Python Code)

Max Activating Token Index: 621

Text #13

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5608. Min Act: -0.0000

Data Index: 1214512 (C4 (Web Text))

Max Activating Token Index: 341

Text #14

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5202. Min Act: -0.0000

Data Index: 957234 (C4 (Web Text))

Max Activating Token Index: 573

Text #15

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5328. Min Act: -0.0000

Data Index: 1358860 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5534. Min Act: -0.0000

Data Index: 1039524 (C4 (Web Text))

Max Activating Token Index: 903

Text #17

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5266. Min Act: -0.0000

Data Index: 11278 (C4 (Web Text))

Max Activating Token Index: 538

Text #18

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5188. Min Act: -0.0000

Data Index: 1225726 (C4 (Web Text))

Max Activating Token Index: 869

Text #19

Max Range: 1.8209. Min Range: -1.8209

Max Act: 1.5637. Min Act: -0.0001

Data Index: 347950 (C4 (Web Text))

Max Activating Token Index: 506