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

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

Neuron 1120 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: 0.0085. Min Range: -0.0085

Max Act: 0.0085. Min Act: -0.0006

Data Index: 1230503 (C4 (Web Text))

Max Activating Token Index: 965

Text #1

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0082. Min Act: -0.0006

Data Index: 427038 (C4 (Web Text))

Max Activating Token Index: 315

Text #2

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0079. Min Act: -0.0006

Data Index: 1104838 (C4 (Web Text))

Max Activating Token Index: 439

Text #3

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0080. Min Act: -0.0006

Data Index: 611821 (C4 (Web Text))

Max Activating Token Index: 409

Text #4

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0077. Min Act: -0.0007

Data Index: 1017478 (C4 (Web Text))

Max Activating Token Index: 765

Text #5

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0074. Min Act: -0.0006

Data Index: 1332179 (C4 (Web Text))

Max Activating Token Index: 335

Text #6

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0077. Min Act: -0.0006

Data Index: 35238 (C4 (Web Text))

Max Activating Token Index: 376

Text #7

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0075. Min Act: -0.0006

Data Index: 210053 (C4 (Web Text))

Max Activating Token Index: 908

Text #8

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0074. Min Act: -0.0006

Data Index: 814474 (C4 (Web Text))

Max Activating Token Index: 395

Text #9

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0074. Min Act: -0.0006

Data Index: 55751 (C4 (Web Text))

Max Activating Token Index: 625

Text #10

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0071. Min Act: -0.0006

Data Index: 858111 (C4 (Web Text))

Max Activating Token Index: 193

Text #11

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0071. Min Act: -0.0006

Data Index: 484563 (C4 (Web Text))

Max Activating Token Index: 748

Text #12

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0070. Min Act: -0.0006

Data Index: 1207689 (C4 (Web Text))

Max Activating Token Index: 273

Text #13

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0070. Min Act: -0.0006

Data Index: 600662 (C4 (Web Text))

Max Activating Token Index: 161

Text #14

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0071. Min Act: -0.0005

Data Index: 1618325 (Python Code)

Max Activating Token Index: 137

Text #15

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0070. Min Act: -0.0007

Data Index: 940735 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0071. Min Act: -0.0006

Data Index: 651778 (C4 (Web Text))

Max Activating Token Index: 58

Text #17

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0070. Min Act: -0.0006

Data Index: 625474 (C4 (Web Text))

Max Activating Token Index: 456

Text #18

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0069. Min Act: -0.0006

Data Index: 1006609 (C4 (Web Text))

Max Activating Token Index: 492

Text #19

Max Range: 0.0085. Min Range: -0.0085

Max Act: 0.0069. Min Act: -0.0006

Data Index: 636900 (C4 (Web Text))

Max Activating Token Index: 450