Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 84 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2298. Min Act: -0.0002

Data Index: 549949 (C4 (Web Text))

Max Activating Token Index: 825

Text #1

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2403. Min Act: -0.0002

Data Index: 1312919 (C4 (Web Text))

Max Activating Token Index: 810

Text #2

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2213. Min Act: -0.0002

Data Index: 1327726 (C4 (Web Text))

Max Activating Token Index: 841

Text #3

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2169. Min Act: -0.0002

Data Index: 1287676 (C4 (Web Text))

Max Activating Token Index: 159

Text #4

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2215. Min Act: -0.0002

Data Index: 1479652 (Python Code)

Max Activating Token Index: 967

Text #5

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1978. Min Act: -0.0002

Data Index: 495587 (C4 (Web Text))

Max Activating Token Index: 785

Text #6

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1941. Min Act: -0.0002

Data Index: 226297 (C4 (Web Text))

Max Activating Token Index: 639

Text #7

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2011. Min Act: -0.0002

Data Index: 708686 (C4 (Web Text))

Max Activating Token Index: 653

Text #8

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1765. Min Act: -0.0002

Data Index: 1602716 (Python Code)

Max Activating Token Index: 821

Text #9

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1977. Min Act: -0.0002

Data Index: 819258 (C4 (Web Text))

Max Activating Token Index: 440

Text #10

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.2230. Min Act: -0.0002

Data Index: 1324352 (C4 (Web Text))

Max Activating Token Index: 722

Text #11

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1945. Min Act: -0.0002

Data Index: 1499441 (Python Code)

Max Activating Token Index: 377

Text #12

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1934. Min Act: -0.0002

Data Index: 792122 (C4 (Web Text))

Max Activating Token Index: 786

Text #13

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1897. Min Act: -0.0002

Data Index: 1212358 (C4 (Web Text))

Max Activating Token Index: 565

Text #14

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1864. Min Act: -0.0002

Data Index: 916938 (C4 (Web Text))

Max Activating Token Index: 763

Text #15

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1814. Min Act: -0.0002

Data Index: 1384267 (Python Code)

Max Activating Token Index: 956

Text #16

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1841. Min Act: -0.0002

Data Index: 1604342 (Python Code)

Max Activating Token Index: 98

Text #17

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1822. Min Act: -0.0002

Data Index: 356336 (C4 (Web Text))

Max Activating Token Index: 908

Text #18

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1937. Min Act: -0.0002

Data Index: 670417 (C4 (Web Text))

Max Activating Token Index: 195

Text #19

Max Range: 0.2403. Min Range: -0.2403

Max Act: 0.1850. Min Act: -0.0002

Data Index: 1303925 (C4 (Web Text))

Max Activating Token Index: 232