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

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

Neuron 3516 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: 4.0237. Min Range: -4.0237

Max Act: 4.0237. Min Act: -0.0000

Data Index: 1440014 (Python Code)

Max Activating Token Index: 504

Text #1

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.1774. Min Act: -0.0000

Data Index: 963461 (C4 (Web Text))

Max Activating Token Index: 927

Text #2

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.2236. Min Act: -0.0000

Data Index: 1400601 (Python Code)

Max Activating Token Index: 346

Text #3

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.1439. Min Act: -0.0000

Data Index: 1581538 (Python Code)

Max Activating Token Index: 38

Text #4

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.0592. Min Act: -0.0000

Data Index: 312788 (C4 (Web Text))

Max Activating Token Index: 883

Text #5

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.0652. Min Act: -0.0000

Data Index: 333259 (C4 (Web Text))

Max Activating Token Index: 716

Text #6

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.0629. Min Act: -0.0000

Data Index: 766257 (C4 (Web Text))

Max Activating Token Index: 821

Text #7

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.0393. Min Act: -0.0000

Data Index: 752638 (C4 (Web Text))

Max Activating Token Index: 935

Text #8

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.9365. Min Act: -0.0000

Data Index: 190226 (C4 (Web Text))

Max Activating Token Index: 304

Text #9

Max Range: 4.0237. Min Range: -4.0237

Max Act: 3.1531. Min Act: -0.0000

Data Index: 1260507 (C4 (Web Text))

Max Activating Token Index: 691

Text #10

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.7240. Min Act: -0.0000

Data Index: 1552684 (Python Code)

Max Activating Token Index: 786

Text #11

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.7193. Min Act: -0.0000

Data Index: 1582692 (Python Code)

Max Activating Token Index: 40

Text #12

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.9727. Min Act: -0.0000

Data Index: 845852 (C4 (Web Text))

Max Activating Token Index: 41

Text #13

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.6942. Min Act: -0.0000

Data Index: 321827 (C4 (Web Text))

Max Activating Token Index: 423

Text #14

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.8276. Min Act: -0.0000

Data Index: 1463557 (Python Code)

Max Activating Token Index: 663

Text #15

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.7887. Min Act: -0.0000

Data Index: 1271591 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.7960. Min Act: -0.0000

Data Index: 101927 (C4 (Web Text))

Max Activating Token Index: 892

Text #17

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.8031. Min Act: -0.0000

Data Index: 1519277 (Python Code)

Max Activating Token Index: 42

Text #18

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.7814. Min Act: -0.0000

Data Index: 1372119 (Python Code)

Max Activating Token Index: 118

Text #19

Max Range: 4.0237. Min Range: -4.0237

Max Act: 2.5086. Min Act: -0.0000

Data Index: 166602 (C4 (Web Text))

Max Activating Token Index: 681