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

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

Neuron 2107 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: 0.5713. Min Range: -0.5713

Max Act: 0.5713. Min Act: -0.0001

Data Index: 133225 (C4 (Web Text))

Max Activating Token Index: 21

Text #1

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.5290. Min Act: -0.0000

Data Index: 541378 (C4 (Web Text))

Max Activating Token Index: 22

Text #2

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.5300. Min Act: -0.0001

Data Index: 115634 (C4 (Web Text))

Max Activating Token Index: 904

Text #3

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.5055. Min Act: -0.0000

Data Index: 272707 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.5146. Min Act: -0.0000

Data Index: 131672 (C4 (Web Text))

Max Activating Token Index: 800

Text #5

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.5131. Min Act: -0.0000

Data Index: 744739 (C4 (Web Text))

Max Activating Token Index: 521

Text #6

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4731. Min Act: -0.0001

Data Index: 609057 (C4 (Web Text))

Max Activating Token Index: 43

Text #7

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4971. Min Act: -0.0001

Data Index: 240309 (C4 (Web Text))

Max Activating Token Index: 72

Text #8

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4618. Min Act: -0.0000

Data Index: 128315 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4945. Min Act: -0.0001

Data Index: 1336042 (C4 (Web Text))

Max Activating Token Index: 774

Text #10

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4924. Min Act: -0.0001

Data Index: 1260345 (C4 (Web Text))

Max Activating Token Index: 99

Text #11

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4244. Min Act: -0.0000

Data Index: 1169981 (C4 (Web Text))

Max Activating Token Index: 180

Text #12

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4532. Min Act: -0.0000

Data Index: 741714 (C4 (Web Text))

Max Activating Token Index: 63

Text #13

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4503. Min Act: -0.0001

Data Index: 532282 (C4 (Web Text))

Max Activating Token Index: 361

Text #14

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4187. Min Act: -0.0001

Data Index: 34960 (C4 (Web Text))

Max Activating Token Index: 35

Text #15

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4317. Min Act: -0.0001

Data Index: 1265405 (C4 (Web Text))

Max Activating Token Index: 253

Text #16

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4203. Min Act: -0.0000

Data Index: 669890 (C4 (Web Text))

Max Activating Token Index: 383

Text #17

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4334. Min Act: -0.0000

Data Index: 1200157 (C4 (Web Text))

Max Activating Token Index: 8

Text #18

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.4158. Min Act: -0.0001

Data Index: 385187 (C4 (Web Text))

Max Activating Token Index: 396

Text #19

Max Range: 0.5713. Min Range: -0.5713

Max Act: 0.3971. Min Act: -0.0000

Data Index: 308439 (C4 (Web Text))

Max Activating Token Index: 839