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

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

Neuron 1832 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.3759. Min Act: -0.0000

Data Index: 808775 (C4 (Web Text))

Max Activating Token Index: 161

Text #1

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.4033. Min Act: -0.0000

Data Index: 189235 (C4 (Web Text))

Max Activating Token Index: 856

Text #2

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.5300. Min Act: -0.0000

Data Index: 1297476 (C4 (Web Text))

Max Activating Token Index: 282

Text #3

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.2804. Min Act: -0.0000

Data Index: 443346 (C4 (Web Text))

Max Activating Token Index: 93

Text #4

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.1512. Min Act: -0.0000

Data Index: 222574 (C4 (Web Text))

Max Activating Token Index: 5

Text #5

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.3704. Min Act: -0.0000

Data Index: 1449512 (Python Code)

Max Activating Token Index: 292

Text #6

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.2548. Min Act: -0.0000

Data Index: 1060520 (C4 (Web Text))

Max Activating Token Index: 55

Text #7

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.2277. Min Act: -0.0000

Data Index: 951600 (C4 (Web Text))

Max Activating Token Index: 747

Text #8

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.2147. Min Act: -0.0000

Data Index: 1650587 (Python Code)

Max Activating Token Index: 406

Text #9

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0907. Min Act: -0.0000

Data Index: 1276966 (C4 (Web Text))

Max Activating Token Index: 975

Text #10

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.1784. Min Act: -0.0000

Data Index: 175681 (C4 (Web Text))

Max Activating Token Index: 642

Text #11

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.1826. Min Act: -0.0000

Data Index: 1568536 (Python Code)

Max Activating Token Index: 73

Text #12

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.1423. Min Act: -0.0000

Data Index: 43077 (C4 (Web Text))

Max Activating Token Index: 11

Text #13

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.1366. Min Act: -0.0000

Data Index: 1404068 (Python Code)

Max Activating Token Index: 165

Text #14

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0988. Min Act: -0.0000

Data Index: 359335 (C4 (Web Text))

Max Activating Token Index: 258

Text #15

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0086. Min Act: -0.0000

Data Index: 1588257 (Python Code)

Max Activating Token Index: 998

Text #16

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0610. Min Act: -0.0000

Data Index: 1321181 (C4 (Web Text))

Max Activating Token Index: 77

Text #17

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0885. Min Act: -0.0000

Data Index: 216319 (C4 (Web Text))

Max Activating Token Index: 270

Text #18

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0922. Min Act: -0.0000

Data Index: 916309 (C4 (Web Text))

Max Activating Token Index: 967

Text #19

Max Range: 2.5300. Min Range: -2.5300

Max Act: 2.0731. Min Act: -0.0000

Data Index: 855108 (C4 (Web Text))

Max Activating Token Index: 9