Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3516 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.3066. Min Act: -0.0001

Data Index: 1056953 (C4 (Web Text))

Max Activating Token Index: 172

Text #1

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2547. Min Act: -0.0000

Data Index: 881453 (C4 (Web Text))

Max Activating Token Index: 754

Text #2

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2248. Min Act: -0.0001

Data Index: 493068 (C4 (Web Text))

Max Activating Token Index: 66

Text #3

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2289. Min Act: -0.0001

Data Index: 273475 (C4 (Web Text))

Max Activating Token Index: 207

Text #4

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2171. Min Act: -0.0000

Data Index: 358206 (C4 (Web Text))

Max Activating Token Index: 479

Text #5

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2044. Min Act: -0.0001

Data Index: 43146 (C4 (Web Text))

Max Activating Token Index: 68

Text #6

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2034. Min Act: -0.0001

Data Index: 119311 (C4 (Web Text))

Max Activating Token Index: 347

Text #7

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2042. Min Act: -0.0000

Data Index: 1345074 (C4 (Web Text))

Max Activating Token Index: 258

Text #8

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2027. Min Act: -0.0001

Data Index: 185514 (C4 (Web Text))

Max Activating Token Index: 335

Text #9

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1955. Min Act: -0.0001

Data Index: 458695 (C4 (Web Text))

Max Activating Token Index: 363

Text #10

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2097. Min Act: -0.0001

Data Index: 1327590 (C4 (Web Text))

Max Activating Token Index: 199

Text #11

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2108. Min Act: -0.0000

Data Index: 364420 (C4 (Web Text))

Max Activating Token Index: 668

Text #12

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2074. Min Act: -0.0001

Data Index: 1404281 (Python Code)

Max Activating Token Index: 196

Text #13

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.2043. Min Act: -0.0001

Data Index: 1336560 (C4 (Web Text))

Max Activating Token Index: 175

Text #14

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1985. Min Act: -0.0000

Data Index: 627929 (C4 (Web Text))

Max Activating Token Index: 215

Text #15

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1851. Min Act: -0.0001

Data Index: 7726 (C4 (Web Text))

Max Activating Token Index: 119

Text #16

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1962. Min Act: -0.0001

Data Index: 1192221 (C4 (Web Text))

Max Activating Token Index: 315

Text #17

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1822. Min Act: -0.0000

Data Index: 90387 (C4 (Web Text))

Max Activating Token Index: 132

Text #18

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1818. Min Act: -0.0000

Data Index: 172993 (C4 (Web Text))

Max Activating Token Index: 434

Text #19

Max Range: 0.3066. Min Range: -0.3066

Max Act: 0.1959. Min Act: -0.0001

Data Index: 1245919 (C4 (Web Text))

Max Activating Token Index: 580