Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4392 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4537. Min Act: -0.0000

Data Index: 294783 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3852. Min Act: -0.0000

Data Index: 322997 (C4 (Web Text))

Max Activating Token Index: 281

Text #2

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3939. Min Act: -0.0000

Data Index: 681811 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3607. Min Act: -0.0000

Data Index: 414102 (C4 (Web Text))

Max Activating Token Index: 935

Text #4

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3862. Min Act: -0.0000

Data Index: 1497313 (Python Code)

Max Activating Token Index: 126

Text #5

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3867. Min Act: -0.0000

Data Index: 289799 (C4 (Web Text))

Max Activating Token Index: 508

Text #6

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3936. Min Act: -0.0000

Data Index: 1596887 (Python Code)

Max Activating Token Index: 726

Text #7

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3638. Min Act: -0.0000

Data Index: 809575 (C4 (Web Text))

Max Activating Token Index: 872

Text #8

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3692. Min Act: -0.0000

Data Index: 68321 (C4 (Web Text))

Max Activating Token Index: 216

Text #9

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3731. Min Act: -0.0000

Data Index: 201562 (C4 (Web Text))

Max Activating Token Index: 263

Text #10

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3313. Min Act: -0.0000

Data Index: 1224328 (C4 (Web Text))

Max Activating Token Index: 520

Text #11

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.2876. Min Act: -0.0000

Data Index: 346291 (C4 (Web Text))

Max Activating Token Index: 276

Text #12

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3459. Min Act: -0.0000

Data Index: 1386837 (Python Code)

Max Activating Token Index: 381

Text #13

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3569. Min Act: -0.0000

Data Index: 801160 (C4 (Web Text))

Max Activating Token Index: 531

Text #14

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3533. Min Act: -0.0000

Data Index: 207943 (C4 (Web Text))

Max Activating Token Index: 416

Text #15

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3416. Min Act: -0.0000

Data Index: 545191 (C4 (Web Text))

Max Activating Token Index: 881

Text #16

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3275. Min Act: -0.0000

Data Index: 1656026 (Python Code)

Max Activating Token Index: 940

Text #17

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3458. Min Act: -0.0000

Data Index: 963490 (C4 (Web Text))

Max Activating Token Index: 646

Text #18

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3174. Min Act: -0.0000

Data Index: 80692 (C4 (Web Text))

Max Activating Token Index: 695

Text #19

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3068. Min Act: -0.0000

Data Index: 760839 (C4 (Web Text))

Max Activating Token Index: 642