Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2264 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3595. Min Act: -0.0001

Data Index: 566621 (C4 (Web Text))

Max Activating Token Index: 141

Text #1

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3560. Min Act: -0.0001

Data Index: 794590 (C4 (Web Text))

Max Activating Token Index: 980

Text #2

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3421. Min Act: -0.0001

Data Index: 896556 (C4 (Web Text))

Max Activating Token Index: 720

Text #3

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3398. Min Act: -0.0001

Data Index: 480540 (C4 (Web Text))

Max Activating Token Index: 589

Text #4

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3438. Min Act: -0.0001

Data Index: 1238991 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3520. Min Act: -0.0001

Data Index: 189455 (C4 (Web Text))

Max Activating Token Index: 694

Text #6

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3319. Min Act: -0.0001

Data Index: 359951 (C4 (Web Text))

Max Activating Token Index: 413

Text #7

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3300. Min Act: -0.0001

Data Index: 639047 (C4 (Web Text))

Max Activating Token Index: 881

Text #8

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3278. Min Act: -0.0001

Data Index: 1092263 (C4 (Web Text))

Max Activating Token Index: 667

Text #9

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3231. Min Act: -0.0001

Data Index: 580400 (C4 (Web Text))

Max Activating Token Index: 727

Text #10

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3215. Min Act: -0.0001

Data Index: 1439759 (Python Code)

Max Activating Token Index: 394

Text #11

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3161. Min Act: -0.0001

Data Index: 395360 (C4 (Web Text))

Max Activating Token Index: 1002

Text #12

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3103. Min Act: -0.0001

Data Index: 1624623 (Python Code)

Max Activating Token Index: 720

Text #13

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3179. Min Act: -0.0001

Data Index: 1251736 (C4 (Web Text))

Max Activating Token Index: 934

Text #14

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3183. Min Act: -0.0001

Data Index: 1340483 (C4 (Web Text))

Max Activating Token Index: 961

Text #15

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3052. Min Act: -0.0001

Data Index: 1478598 (Python Code)

Max Activating Token Index: 36

Text #16

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3061. Min Act: -0.0001

Data Index: 185331 (C4 (Web Text))

Max Activating Token Index: 146

Text #17

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.3010. Min Act: -0.0001

Data Index: 1547094 (Python Code)

Max Activating Token Index: 112

Text #18

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.2817. Min Act: -0.0001

Data Index: 903956 (C4 (Web Text))

Max Activating Token Index: 357

Text #19

Max Range: 0.3595. Min Range: -0.3595

Max Act: 0.2954. Min Act: -0.0001

Data Index: 850819 (C4 (Web Text))

Max Activating Token Index: 968