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

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

Neuron 2968 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.4084. Min Act: -0.0000

Data Index: 757325 (C4 (Web Text))

Max Activating Token Index: 23

Text #1

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.3119. Min Act: -0.0000

Data Index: 1531269 (Python Code)

Max Activating Token Index: 764

Text #2

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2849. Min Act: -0.0001

Data Index: 1585153 (Python Code)

Max Activating Token Index: 470

Text #3

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2842. Min Act: -0.0001

Data Index: 1467726 (Python Code)

Max Activating Token Index: 92

Text #4

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2785. Min Act: -0.0000

Data Index: 866331 (C4 (Web Text))

Max Activating Token Index: 793

Text #5

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.3072. Min Act: -0.0001

Data Index: 1228177 (C4 (Web Text))

Max Activating Token Index: 775

Text #6

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2623. Min Act: -0.0001

Data Index: 975406 (C4 (Web Text))

Max Activating Token Index: 588

Text #7

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2430. Min Act: -0.0001

Data Index: 1523452 (Python Code)

Max Activating Token Index: 963

Text #8

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2341. Min Act: -0.0001

Data Index: 223284 (C4 (Web Text))

Max Activating Token Index: 260

Text #9

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2328. Min Act: -0.0000

Data Index: 590074 (C4 (Web Text))

Max Activating Token Index: 746

Text #10

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2477. Min Act: -0.0001

Data Index: 628067 (C4 (Web Text))

Max Activating Token Index: 975

Text #11

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2221. Min Act: -0.0001

Data Index: 1386893 (Python Code)

Max Activating Token Index: 422

Text #12

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2356. Min Act: -0.0001

Data Index: 1627142 (Python Code)

Max Activating Token Index: 657

Text #13

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2232. Min Act: -0.0001

Data Index: 1607178 (Python Code)

Max Activating Token Index: 365

Text #14

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2212. Min Act: -0.0000

Data Index: 1602089 (Python Code)

Max Activating Token Index: 927

Text #15

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2282. Min Act: -0.0001

Data Index: 1258757 (C4 (Web Text))

Max Activating Token Index: 363

Text #16

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2110. Min Act: -0.0001

Data Index: 720789 (C4 (Web Text))

Max Activating Token Index: 878

Text #17

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2271. Min Act: -0.0001

Data Index: 1646750 (Python Code)

Max Activating Token Index: 299

Text #18

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2244. Min Act: -0.0000

Data Index: 1343394 (C4 (Web Text))

Max Activating Token Index: 979

Text #19

Max Range: 0.4084. Min Range: -0.4084

Max Act: 0.2156. Min Act: -0.0001

Data Index: 1605107 (Python Code)

Max Activating Token Index: 439