Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 2021 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0148. Min Act: -0.0005

Data Index: 1288927 (C4 (Web Text))

Max Activating Token Index: 205

Text #1

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0155. Min Act: -0.0005

Data Index: 840485 (C4 (Web Text))

Max Activating Token Index: 857

Text #2

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0148. Min Act: -0.0005

Data Index: 176387 (C4 (Web Text))

Max Activating Token Index: 754

Text #3

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0142. Min Act: -0.0005

Data Index: 268430 (C4 (Web Text))

Max Activating Token Index: 747

Text #4

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0147. Min Act: -0.0005

Data Index: 790309 (C4 (Web Text))

Max Activating Token Index: 397

Text #5

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0143. Min Act: -0.0005

Data Index: 1264478 (C4 (Web Text))

Max Activating Token Index: 706

Text #6

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0135. Min Act: -0.0005

Data Index: 663303 (C4 (Web Text))

Max Activating Token Index: 113

Text #7

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0134. Min Act: -0.0005

Data Index: 689105 (C4 (Web Text))

Max Activating Token Index: 976

Text #8

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0131. Min Act: -0.0005

Data Index: 213646 (C4 (Web Text))

Max Activating Token Index: 991

Text #9

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0134. Min Act: -0.0005

Data Index: 625858 (C4 (Web Text))

Max Activating Token Index: 164

Text #10

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0121. Min Act: -0.0005

Data Index: 610662 (C4 (Web Text))

Max Activating Token Index: 254

Text #11

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0132. Min Act: -0.0005

Data Index: 473827 (C4 (Web Text))

Max Activating Token Index: 378

Text #12

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0132. Min Act: -0.0005

Data Index: 764840 (C4 (Web Text))

Max Activating Token Index: 368

Text #13

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0131. Min Act: -0.0005

Data Index: 1026177 (C4 (Web Text))

Max Activating Token Index: 158

Text #14

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0128. Min Act: -0.0005

Data Index: 217037 (C4 (Web Text))

Max Activating Token Index: 778

Text #15

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0124. Min Act: -0.0005

Data Index: 299299 (C4 (Web Text))

Max Activating Token Index: 312

Text #16

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0127. Min Act: -0.0005

Data Index: 479577 (C4 (Web Text))

Max Activating Token Index: 563

Text #17

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0124. Min Act: -0.0005

Data Index: 1172881 (C4 (Web Text))

Max Activating Token Index: 153

Text #18

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0120. Min Act: -0.0005

Data Index: 1585199 (Python Code)

Max Activating Token Index: 874

Text #19

Max Range: 0.0155. Min Range: -0.0155

Max Act: 0.0129. Min Act: -0.0005

Data Index: 324676 (C4 (Web Text))

Max Activating Token Index: 707