Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2010 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0241. Min Act: -0.0006

Data Index: 786052 (C4 (Web Text))

Max Activating Token Index: 53

Text #1

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0219. Min Act: -0.0006

Data Index: 645814 (C4 (Web Text))

Max Activating Token Index: 683

Text #2

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0209. Min Act: -0.0006

Data Index: 249522 (C4 (Web Text))

Max Activating Token Index: 505

Text #3

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0203. Min Act: -0.0006

Data Index: 749423 (C4 (Web Text))

Max Activating Token Index: 83

Text #4

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0201. Min Act: -0.0005

Data Index: 1077729 (C4 (Web Text))

Max Activating Token Index: 710

Text #5

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0196. Min Act: -0.0006

Data Index: 775103 (C4 (Web Text))

Max Activating Token Index: 164

Text #6

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0191. Min Act: -0.0006

Data Index: 483632 (C4 (Web Text))

Max Activating Token Index: 935

Text #7

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0192. Min Act: -0.0006

Data Index: 391491 (C4 (Web Text))

Max Activating Token Index: 741

Text #8

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0187. Min Act: -0.0006

Data Index: 50042 (C4 (Web Text))

Max Activating Token Index: 908

Text #9

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0196. Min Act: -0.0006

Data Index: 856419 (C4 (Web Text))

Max Activating Token Index: 247

Text #10

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0195. Min Act: -0.0005

Data Index: 112236 (C4 (Web Text))

Max Activating Token Index: 119

Text #11

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0190. Min Act: -0.0006

Data Index: 374099 (C4 (Web Text))

Max Activating Token Index: 676

Text #12

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0185. Min Act: -0.0006

Data Index: 1137364 (C4 (Web Text))

Max Activating Token Index: 892

Text #13

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0189. Min Act: -0.0005

Data Index: 455465 (C4 (Web Text))

Max Activating Token Index: 75

Text #14

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0189. Min Act: -0.0006

Data Index: 899859 (C4 (Web Text))

Max Activating Token Index: 348

Text #15

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0188. Min Act: -0.0006

Data Index: 554640 (C4 (Web Text))

Max Activating Token Index: 752

Text #16

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0187. Min Act: -0.0006

Data Index: 1010021 (C4 (Web Text))

Max Activating Token Index: 526

Text #17

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0185. Min Act: -0.0006

Data Index: 841659 (C4 (Web Text))

Max Activating Token Index: 406

Text #18

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0185. Min Act: -0.0005

Data Index: 562412 (C4 (Web Text))

Max Activating Token Index: 649

Text #19

Max Range: 0.0241. Min Range: -0.0241

Max Act: 0.0185. Min Act: -0.0006

Data Index: 1020360 (C4 (Web Text))

Max Activating Token Index: 555