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

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

Neuron 2581 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: 0.0218. Min Range: -0.0218

Max Act: 0.0157. Min Act: -0.0000

Data Index: 650220 (C4 (Web Text))

Max Activating Token Index: 42

Text #1

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0167. Min Act: -0.0000

Data Index: 272448 (C4 (Web Text))

Max Activating Token Index: 327

Text #2

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0218. Min Act: -0.0000

Data Index: 361369 (C4 (Web Text))

Max Activating Token Index: 188

Text #3

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0180. Min Act: -0.0000

Data Index: 727254 (C4 (Web Text))

Max Activating Token Index: 386

Text #4

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0170. Min Act: -0.0000

Data Index: 838371 (C4 (Web Text))

Max Activating Token Index: 733

Text #5

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0198. Min Act: -0.0000

Data Index: 694802 (C4 (Web Text))

Max Activating Token Index: 747

Text #6

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0192. Min Act: -0.0001

Data Index: 176687 (C4 (Web Text))

Max Activating Token Index: 298

Text #7

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0161. Min Act: -0.0001

Data Index: 603190 (C4 (Web Text))

Max Activating Token Index: 261

Text #8

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0204. Min Act: -0.0000

Data Index: 877066 (C4 (Web Text))

Max Activating Token Index: 349

Text #9

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0168. Min Act: -0.0000

Data Index: 294873 (C4 (Web Text))

Max Activating Token Index: 93

Text #10

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0160. Min Act: -0.0000

Data Index: 551953 (C4 (Web Text))

Max Activating Token Index: 840

Text #11

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0160. Min Act: -0.0000

Data Index: 534820 (C4 (Web Text))

Max Activating Token Index: 638

Text #12

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0157. Min Act: -0.0000

Data Index: 1011947 (C4 (Web Text))

Max Activating Token Index: 125

Text #13

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0162. Min Act: -0.0000

Data Index: 858478 (C4 (Web Text))

Max Activating Token Index: 831

Text #14

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0138. Min Act: -0.0000

Data Index: 1408915 (Python Code)

Max Activating Token Index: 203

Text #15

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0147. Min Act: -0.0000

Data Index: 1199262 (C4 (Web Text))

Max Activating Token Index: 93

Text #16

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0190. Min Act: -0.0000

Data Index: 724074 (C4 (Web Text))

Max Activating Token Index: 729

Text #17

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0093. Min Act: -0.0000

Data Index: 920766 (C4 (Web Text))

Max Activating Token Index: 146

Text #18

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0160. Min Act: -0.0000

Data Index: 892650 (C4 (Web Text))

Max Activating Token Index: 713

Text #19

Max Range: 0.0218. Min Range: -0.0218

Max Act: 0.0160. Min Act: -0.0001

Data Index: 786390 (C4 (Web Text))

Max Activating Token Index: 659