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

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

Neuron 3375 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.3545. Min Act: -0.0001

Data Index: 295072 (C4 (Web Text))

Max Activating Token Index: 600

Text #1

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.3066. Min Act: -0.0001

Data Index: 602519 (C4 (Web Text))

Max Activating Token Index: 967

Text #2

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2920. Min Act: -0.0001

Data Index: 837681 (C4 (Web Text))

Max Activating Token Index: 30

Text #3

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2968. Min Act: -0.0001

Data Index: 750281 (C4 (Web Text))

Max Activating Token Index: 47

Text #4

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2807. Min Act: -0.0001

Data Index: 236338 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2554. Min Act: -0.0001

Data Index: 180207 (C4 (Web Text))

Max Activating Token Index: 318

Text #6

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2614. Min Act: -0.0000

Data Index: 95944 (C4 (Web Text))

Max Activating Token Index: 323

Text #7

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2761. Min Act: -0.0001

Data Index: 147916 (C4 (Web Text))

Max Activating Token Index: 123

Text #8

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2487. Min Act: -0.0001

Data Index: 323785 (C4 (Web Text))

Max Activating Token Index: 94

Text #9

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2503. Min Act: -0.0000

Data Index: 507935 (C4 (Web Text))

Max Activating Token Index: 132

Text #10

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2546. Min Act: -0.0001

Data Index: 649504 (C4 (Web Text))

Max Activating Token Index: 37

Text #11

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2474. Min Act: -0.0001

Data Index: 302462 (C4 (Web Text))

Max Activating Token Index: 64

Text #12

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2402. Min Act: -0.0001

Data Index: 306898 (C4 (Web Text))

Max Activating Token Index: 338

Text #13

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2171. Min Act: -0.0000

Data Index: 995425 (C4 (Web Text))

Max Activating Token Index: 769

Text #14

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2382. Min Act: -0.0001

Data Index: 556269 (C4 (Web Text))

Max Activating Token Index: 576

Text #15

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2380. Min Act: -0.0001

Data Index: 26417 (C4 (Web Text))

Max Activating Token Index: 169

Text #16

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2179. Min Act: -0.0001

Data Index: 1066699 (C4 (Web Text))

Max Activating Token Index: 208

Text #17

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2234. Min Act: -0.0001

Data Index: 420701 (C4 (Web Text))

Max Activating Token Index: 520

Text #18

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.2173. Min Act: -0.0001

Data Index: 1015894 (C4 (Web Text))

Max Activating Token Index: 78

Text #19

Max Range: 0.3545. Min Range: -0.3545

Max Act: 0.1979. Min Act: -0.0001

Data Index: 96388 (C4 (Web Text))

Max Activating Token Index: 286