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

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

Neuron 3475 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.4637. Min Act: -0.0000

Data Index: 583153 (C4 (Web Text))

Max Activating Token Index: 685

Text #1

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.2445. Min Act: -0.0001

Data Index: 947120 (C4 (Web Text))

Max Activating Token Index: 335

Text #2

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.2394. Min Act: -0.0000

Data Index: 1427099 (Python Code)

Max Activating Token Index: 191

Text #3

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.2131. Min Act: -0.0000

Data Index: 488694 (C4 (Web Text))

Max Activating Token Index: 154

Text #4

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0545. Min Act: -0.0000

Data Index: 948166 (C4 (Web Text))

Max Activating Token Index: 315

Text #5

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0590. Min Act: -0.0001

Data Index: 166935 (C4 (Web Text))

Max Activating Token Index: 497

Text #6

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0513. Min Act: -0.0000

Data Index: 445200 (C4 (Web Text))

Max Activating Token Index: 971

Text #7

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.2138. Min Act: -0.0000

Data Index: 741087 (C4 (Web Text))

Max Activating Token Index: 445

Text #8

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.9432. Min Act: -0.0000

Data Index: 198443 (C4 (Web Text))

Max Activating Token Index: 922

Text #9

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.8323. Min Act: -0.0000

Data Index: 332172 (C4 (Web Text))

Max Activating Token Index: 311

Text #10

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.9057. Min Act: -0.0000

Data Index: 1449344 (Python Code)

Max Activating Token Index: 202

Text #11

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0104. Min Act: -0.0001

Data Index: 1273467 (C4 (Web Text))

Max Activating Token Index: 346

Text #12

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0012. Min Act: -0.0001

Data Index: 840204 (C4 (Web Text))

Max Activating Token Index: 902

Text #13

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0685. Min Act: -0.0001

Data Index: 979054 (C4 (Web Text))

Max Activating Token Index: 973

Text #14

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.9730. Min Act: -0.0001

Data Index: 1257525 (C4 (Web Text))

Max Activating Token Index: 956

Text #15

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.8649. Min Act: -0.0001

Data Index: 115716 (C4 (Web Text))

Max Activating Token Index: 541

Text #16

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.9479. Min Act: -0.0001

Data Index: 913091 (C4 (Web Text))

Max Activating Token Index: 1010

Text #17

Max Range: 2.4637. Min Range: -2.4637

Max Act: 2.0089. Min Act: -0.0000

Data Index: 1449545 (Python Code)

Max Activating Token Index: 717

Text #18

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.8174. Min Act: -0.0000

Data Index: 1636696 (Python Code)

Max Activating Token Index: 885

Text #19

Max Range: 2.4637. Min Range: -2.4637

Max Act: 1.9955. Min Act: -0.0000

Data Index: 368305 (C4 (Web Text))

Max Activating Token Index: 623