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

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

Neuron 4814 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.5306. Min Act: -0.0002

Data Index: 1583376 (Python Code)

Max Activating Token Index: 528

Text #1

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4985. Min Act: -0.0002

Data Index: 1017051 (C4 (Web Text))

Max Activating Token Index: 334

Text #2

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4895. Min Act: -0.0002

Data Index: 443526 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4683. Min Act: -0.0001

Data Index: 1168048 (C4 (Web Text))

Max Activating Token Index: 766

Text #4

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4759. Min Act: -0.0001

Data Index: 1476070 (Python Code)

Max Activating Token Index: 562

Text #5

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4445. Min Act: -0.0002

Data Index: 55351 (C4 (Web Text))

Max Activating Token Index: 189

Text #6

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4228. Min Act: -0.0002

Data Index: 219890 (C4 (Web Text))

Max Activating Token Index: 895

Text #7

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4377. Min Act: -0.0002

Data Index: 511793 (C4 (Web Text))

Max Activating Token Index: 802

Text #8

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.4048. Min Act: -0.0001

Data Index: 956887 (C4 (Web Text))

Max Activating Token Index: 186

Text #9

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3942. Min Act: -0.0001

Data Index: 1527187 (Python Code)

Max Activating Token Index: 429

Text #10

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3859. Min Act: -0.0001

Data Index: 1620131 (Python Code)

Max Activating Token Index: 632

Text #11

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3756. Min Act: -0.0001

Data Index: 1510989 (Python Code)

Max Activating Token Index: 276

Text #12

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3761. Min Act: -0.0002

Data Index: 1371363 (Python Code)

Max Activating Token Index: 521

Text #13

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3440. Min Act: -0.0002

Data Index: 679208 (C4 (Web Text))

Max Activating Token Index: 740

Text #14

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3491. Min Act: -0.0002

Data Index: 675741 (C4 (Web Text))

Max Activating Token Index: 623

Text #15

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3487. Min Act: -0.0001

Data Index: 1507179 (Python Code)

Max Activating Token Index: 780

Text #16

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3462. Min Act: -0.0002

Data Index: 499637 (C4 (Web Text))

Max Activating Token Index: 481

Text #17

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3546. Min Act: -0.0001

Data Index: 799116 (C4 (Web Text))

Max Activating Token Index: 868

Text #18

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3568. Min Act: -0.0002

Data Index: 243782 (C4 (Web Text))

Max Activating Token Index: 496

Text #19

Max Range: 0.5306. Min Range: -0.5306

Max Act: 0.3338. Min Act: -0.0002

Data Index: 896896 (C4 (Web Text))

Max Activating Token Index: 667