Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2384 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.8628. Min Act: -0.0003

Data Index: 619939 (C4 (Web Text))

Max Activating Token Index: 137

Text #1

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.8777. Min Act: -0.0003

Data Index: 83489 (C4 (Web Text))

Max Activating Token Index: 301

Text #2

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7950. Min Act: -0.0003

Data Index: 708705 (C4 (Web Text))

Max Activating Token Index: 232

Text #3

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7989. Min Act: -0.0003

Data Index: 481115 (C4 (Web Text))

Max Activating Token Index: 987

Text #4

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7662. Min Act: -0.0003

Data Index: 867721 (C4 (Web Text))

Max Activating Token Index: 199

Text #5

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7813. Min Act: -0.0003

Data Index: 381852 (C4 (Web Text))

Max Activating Token Index: 533

Text #6

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7487. Min Act: -0.0003

Data Index: 966815 (C4 (Web Text))

Max Activating Token Index: 547

Text #7

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7999. Min Act: -0.0003

Data Index: 813581 (C4 (Web Text))

Max Activating Token Index: 49

Text #8

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7643. Min Act: -0.0003

Data Index: 214326 (C4 (Web Text))

Max Activating Token Index: 660

Text #9

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.9440. Min Act: -0.0003

Data Index: 1025444 (C4 (Web Text))

Max Activating Token Index: 615

Text #10

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7435. Min Act: -0.0003

Data Index: 943156 (C4 (Web Text))

Max Activating Token Index: 439

Text #11

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7355. Min Act: -0.0003

Data Index: 1426 (C4 (Web Text))

Max Activating Token Index: 275

Text #12

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7514. Min Act: -0.0003

Data Index: 429800 (C4 (Web Text))

Max Activating Token Index: 815

Text #13

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.8430. Min Act: -0.0003

Data Index: 719696 (C4 (Web Text))

Max Activating Token Index: 623

Text #14

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.6962. Min Act: -0.0003

Data Index: 780119 (C4 (Web Text))

Max Activating Token Index: 278

Text #15

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7595. Min Act: -0.0003

Data Index: 578714 (C4 (Web Text))

Max Activating Token Index: 279

Text #16

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7101. Min Act: -0.0003

Data Index: 1133804 (C4 (Web Text))

Max Activating Token Index: 489

Text #17

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.6167. Min Act: -0.0003

Data Index: 814255 (C4 (Web Text))

Max Activating Token Index: 532

Text #18

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.6690. Min Act: -0.0003

Data Index: 1223402 (C4 (Web Text))

Max Activating Token Index: 276

Text #19

Max Range: 0.9440. Min Range: -0.9440

Max Act: 0.7345. Min Act: -0.0003

Data Index: 576218 (C4 (Web Text))

Max Activating Token Index: 370