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

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

Neuron 805 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: 0.0164. Min Range: -0.0164

Max Act: 0.0164. Min Act: -0.0005

Data Index: 1536182 (Python Code)

Max Activating Token Index: 665

Text #1

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0153. Min Act: -0.0006

Data Index: 106507 (C4 (Web Text))

Max Activating Token Index: 837

Text #2

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0148. Min Act: -0.0006

Data Index: 329463 (C4 (Web Text))

Max Activating Token Index: 943

Text #3

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0144. Min Act: -0.0005

Data Index: 1583000 (Python Code)

Max Activating Token Index: 261

Text #4

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0143. Min Act: -0.0005

Data Index: 1545755 (Python Code)

Max Activating Token Index: 311

Text #5

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0139. Min Act: -0.0005

Data Index: 1415805 (Python Code)

Max Activating Token Index: 843

Text #6

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0139. Min Act: -0.0006

Data Index: 340446 (C4 (Web Text))

Max Activating Token Index: 368

Text #7

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0139. Min Act: -0.0006

Data Index: 317374 (C4 (Web Text))

Max Activating Token Index: 621

Text #8

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0138. Min Act: -0.0005

Data Index: 488265 (C4 (Web Text))

Max Activating Token Index: 491

Text #9

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0136. Min Act: -0.0006

Data Index: 344213 (C4 (Web Text))

Max Activating Token Index: 311

Text #10

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0135. Min Act: -0.0006

Data Index: 894948 (C4 (Web Text))

Max Activating Token Index: 381

Text #11

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0128. Min Act: -0.0006

Data Index: 472345 (C4 (Web Text))

Max Activating Token Index: 662

Text #12

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0126. Min Act: -0.0006

Data Index: 800265 (C4 (Web Text))

Max Activating Token Index: 261

Text #13

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0128. Min Act: -0.0005

Data Index: 1393082 (Python Code)

Max Activating Token Index: 21

Text #14

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0128. Min Act: -0.0006

Data Index: 54375 (C4 (Web Text))

Max Activating Token Index: 594

Text #15

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0128. Min Act: -0.0006

Data Index: 1285656 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0122. Min Act: -0.0006

Data Index: 1420756 (Python Code)

Max Activating Token Index: 646

Text #17

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0121. Min Act: -0.0006

Data Index: 1006602 (C4 (Web Text))

Max Activating Token Index: 21

Text #18

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0124. Min Act: -0.0006

Data Index: 1445021 (Python Code)

Max Activating Token Index: 122

Text #19

Max Range: 0.0164. Min Range: -0.0164

Max Act: 0.0124. Min Act: -0.0006

Data Index: 888421 (C4 (Web Text))

Max Activating Token Index: 208