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

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

Neuron 3072 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: 0.3191. Min Range: -0.3191

Max Act: 0.3191. Min Act: -0.0001

Data Index: 1394413 (Python Code)

Max Activating Token Index: 895

Text #1

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2535. Min Act: -0.0001

Data Index: 1504407 (Python Code)

Max Activating Token Index: 343

Text #2

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.3100. Min Act: -0.0001

Data Index: 136109 (C4 (Web Text))

Max Activating Token Index: 19

Text #3

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2731. Min Act: -0.0000

Data Index: 1422113 (Python Code)

Max Activating Token Index: 181

Text #4

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2395. Min Act: -0.0001

Data Index: 1479112 (Python Code)

Max Activating Token Index: 575

Text #5

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2332. Min Act: -0.0001

Data Index: 1429526 (Python Code)

Max Activating Token Index: 562

Text #6

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2140. Min Act: -0.0001

Data Index: 906069 (C4 (Web Text))

Max Activating Token Index: 949

Text #7

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2280. Min Act: -0.0001

Data Index: 1602771 (Python Code)

Max Activating Token Index: 889

Text #8

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2157. Min Act: -0.0000

Data Index: 1502226 (Python Code)

Max Activating Token Index: 417

Text #9

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2166. Min Act: -0.0000

Data Index: 1645517 (Python Code)

Max Activating Token Index: 414

Text #10

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2170. Min Act: -0.0000

Data Index: 1494226 (Python Code)

Max Activating Token Index: 301

Text #11

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2244. Min Act: -0.0001

Data Index: 1564077 (Python Code)

Max Activating Token Index: 227

Text #12

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2015. Min Act: -0.0001

Data Index: 1287929 (C4 (Web Text))

Max Activating Token Index: 318

Text #13

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2195. Min Act: -0.0001

Data Index: 1509155 (Python Code)

Max Activating Token Index: 723

Text #14

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2100. Min Act: -0.0001

Data Index: 1651667 (Python Code)

Max Activating Token Index: 250

Text #15

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2003. Min Act: -0.0001

Data Index: 550799 (C4 (Web Text))

Max Activating Token Index: 76

Text #16

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2066. Min Act: -0.0000

Data Index: 494625 (C4 (Web Text))

Max Activating Token Index: 398

Text #17

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.2223. Min Act: -0.0001

Data Index: 1495659 (Python Code)

Max Activating Token Index: 744

Text #18

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.1924. Min Act: -0.0001

Data Index: 1373084 (Python Code)

Max Activating Token Index: 33

Text #19

Max Range: 0.3191. Min Range: -0.3191

Max Act: 0.1830. Min Act: -0.0000

Data Index: 1493839 (Python Code)

Max Activating Token Index: 129