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

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

Neuron 4580 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.9598. Min Act: -0.0001

Data Index: 197929 (C4 (Web Text))

Max Activating Token Index: 929

Text #1

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.9102. Min Act: -0.0001

Data Index: 1165273 (C4 (Web Text))

Max Activating Token Index: 428

Text #2

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.7058. Min Act: -0.0001

Data Index: 785036 (C4 (Web Text))

Max Activating Token Index: 238

Text #3

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.9275. Min Act: -0.0000

Data Index: 386320 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.8655. Min Act: -0.0001

Data Index: 1149854 (C4 (Web Text))

Max Activating Token Index: 104

Text #5

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.6856. Min Act: -0.0001

Data Index: 373772 (C4 (Web Text))

Max Activating Token Index: 224

Text #6

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.6134. Min Act: -0.0001

Data Index: 196752 (C4 (Web Text))

Max Activating Token Index: 786

Text #7

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.6519. Min Act: -0.0000

Data Index: 1484553 (Python Code)

Max Activating Token Index: 865

Text #8

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.6021. Min Act: -0.0001

Data Index: 1281929 (C4 (Web Text))

Max Activating Token Index: 154

Text #9

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.4748. Min Act: -0.0000

Data Index: 1386367 (Python Code)

Max Activating Token Index: 300

Text #10

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.6602. Min Act: -0.0000

Data Index: 216252 (C4 (Web Text))

Max Activating Token Index: 500

Text #11

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.4863. Min Act: -0.0001

Data Index: 1049961 (C4 (Web Text))

Max Activating Token Index: 842

Text #12

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.5956. Min Act: -0.0000

Data Index: 1612746 (Python Code)

Max Activating Token Index: 359

Text #13

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.5130. Min Act: -0.0000

Data Index: 1308600 (C4 (Web Text))

Max Activating Token Index: 711

Text #14

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.3907. Min Act: -0.0000

Data Index: 1372552 (Python Code)

Max Activating Token Index: 53

Text #15

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.5321. Min Act: -0.0000

Data Index: 112182 (C4 (Web Text))

Max Activating Token Index: 668

Text #16

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.4463. Min Act: -0.0001

Data Index: 678997 (C4 (Web Text))

Max Activating Token Index: 270

Text #17

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.4351. Min Act: -0.0000

Data Index: 1369735 (Python Code)

Max Activating Token Index: 416

Text #18

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.5598. Min Act: -0.0000

Data Index: 592034 (C4 (Web Text))

Max Activating Token Index: 734

Text #19

Max Range: 1.9598. Min Range: -1.9598

Max Act: 1.3544. Min Act: -0.0000

Data Index: 1404994 (Python Code)

Max Activating Token Index: 295