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

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

Neuron 3765 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6939. Min Act: -0.0002

Data Index: 1040948 (C4 (Web Text))

Max Activating Token Index: 526

Text #1

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6861. Min Act: -0.0002

Data Index: 385398 (C4 (Web Text))

Max Activating Token Index: 486

Text #2

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6517. Min Act: -0.0002

Data Index: 1127424 (C4 (Web Text))

Max Activating Token Index: 496

Text #3

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6635. Min Act: -0.0002

Data Index: 1175988 (C4 (Web Text))

Max Activating Token Index: 124

Text #4

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6653. Min Act: -0.0003

Data Index: 484450 (C4 (Web Text))

Max Activating Token Index: 814

Text #5

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6463. Min Act: -0.0003

Data Index: 381595 (C4 (Web Text))

Max Activating Token Index: 601

Text #6

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6391. Min Act: -0.0003

Data Index: 1147223 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6226. Min Act: -0.0003

Data Index: 669064 (C4 (Web Text))

Max Activating Token Index: 232

Text #8

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6429. Min Act: -0.0003

Data Index: 843330 (C4 (Web Text))

Max Activating Token Index: 199

Text #9

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6208. Min Act: -0.0003

Data Index: 347575 (C4 (Web Text))

Max Activating Token Index: 451

Text #10

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6167. Min Act: -0.0003

Data Index: 587224 (C4 (Web Text))

Max Activating Token Index: 157

Text #11

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6104. Min Act: -0.0003

Data Index: 48830 (C4 (Web Text))

Max Activating Token Index: 117

Text #12

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6057. Min Act: -0.0003

Data Index: 289900 (C4 (Web Text))

Max Activating Token Index: 581

Text #13

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6023. Min Act: -0.0003

Data Index: 953660 (C4 (Web Text))

Max Activating Token Index: 342

Text #14

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.6020. Min Act: -0.0003

Data Index: 284762 (C4 (Web Text))

Max Activating Token Index: 252

Text #15

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.5776. Min Act: -0.0003

Data Index: 244245 (C4 (Web Text))

Max Activating Token Index: 798

Text #16

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.5811. Min Act: -0.0003

Data Index: 1155165 (C4 (Web Text))

Max Activating Token Index: 779

Text #17

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.5780. Min Act: -0.0003

Data Index: 99865 (C4 (Web Text))

Max Activating Token Index: 691

Text #18

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.5673. Min Act: -0.0003

Data Index: 194244 (C4 (Web Text))

Max Activating Token Index: 353

Text #19

Max Range: 0.6939. Min Range: -0.6939

Max Act: 0.5625. Min Act: -0.0003

Data Index: 84185 (C4 (Web Text))

Max Activating Token Index: 613