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

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

Neuron 4210 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3933. Min Range: -1.3933

Max Act: 1.3933. Min Act: -0.0001

Data Index: 798764 (C4 (Web Text))

Max Activating Token Index: 143

Text #1

Max Range: 1.3933. Min Range: -1.3933

Max Act: 1.3545. Min Act: -0.0001

Data Index: 1102171 (C4 (Web Text))

Max Activating Token Index: 470

Text #2

Max Range: 1.3933. Min Range: -1.3933

Max Act: 1.3809. Min Act: -0.0001

Data Index: 1144891 (C4 (Web Text))

Max Activating Token Index: 398

Text #3

Max Range: 1.3933. Min Range: -1.3933

Max Act: 1.2204. Min Act: -0.0001

Data Index: 375536 (C4 (Web Text))

Max Activating Token Index: 26

Text #4

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.9536. Min Act: -0.0001

Data Index: 1127419 (C4 (Web Text))

Max Activating Token Index: 9

Text #5

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.9437. Min Act: -0.0001

Data Index: 611589 (C4 (Web Text))

Max Activating Token Index: 505

Text #6

Max Range: 1.3933. Min Range: -1.3933

Max Act: 1.0149. Min Act: -0.0001

Data Index: 1038721 (C4 (Web Text))

Max Activating Token Index: 900

Text #7

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8993. Min Act: -0.0001

Data Index: 967359 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.9068. Min Act: -0.0001

Data Index: 358564 (C4 (Web Text))

Max Activating Token Index: 5

Text #9

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.9171. Min Act: -0.0001

Data Index: 157640 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8618. Min Act: -0.0001

Data Index: 163131 (C4 (Web Text))

Max Activating Token Index: 504

Text #11

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8673. Min Act: -0.0001

Data Index: 990980 (C4 (Web Text))

Max Activating Token Index: 201

Text #12

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8043. Min Act: -0.0001

Data Index: 912861 (C4 (Web Text))

Max Activating Token Index: 652

Text #13

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.9180. Min Act: -0.0001

Data Index: 1006784 (C4 (Web Text))

Max Activating Token Index: 284

Text #14

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8320. Min Act: -0.0001

Data Index: 502431 (C4 (Web Text))

Max Activating Token Index: 141

Text #15

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8736. Min Act: -0.0001

Data Index: 573595 (C4 (Web Text))

Max Activating Token Index: 318

Text #16

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8541. Min Act: -0.0001

Data Index: 283568 (C4 (Web Text))

Max Activating Token Index: 658

Text #17

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8001. Min Act: -0.0001

Data Index: 388718 (C4 (Web Text))

Max Activating Token Index: 37

Text #18

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.8375. Min Act: -0.0001

Data Index: 65714 (C4 (Web Text))

Max Activating Token Index: 136

Text #19

Max Range: 1.3933. Min Range: -1.3933

Max Act: 0.7537. Min Act: -0.0001

Data Index: 1188216 (C4 (Web Text))

Max Activating Token Index: 542