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

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

Neuron 560 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: 0.3847. Min Range: -0.3847

Max Act: 0.3847. Min Act: -0.0001

Data Index: 1296531 (C4 (Web Text))

Max Activating Token Index: 235

Text #1

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1851. Min Act: -0.0001

Data Index: 865033 (C4 (Web Text))

Max Activating Token Index: 679

Text #2

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.2971. Min Act: -0.0001

Data Index: 369921 (C4 (Web Text))

Max Activating Token Index: 92

Text #3

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.2179. Min Act: -0.0001

Data Index: 510136 (C4 (Web Text))

Max Activating Token Index: 135

Text #4

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.2402. Min Act: -0.0001

Data Index: 160119 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.2296. Min Act: -0.0001

Data Index: 610658 (C4 (Web Text))

Max Activating Token Index: 359

Text #6

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1845. Min Act: -0.0001

Data Index: 585759 (C4 (Web Text))

Max Activating Token Index: 276

Text #7

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1227. Min Act: -0.0000

Data Index: 134749 (C4 (Web Text))

Max Activating Token Index: 907

Text #8

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1884. Min Act: -0.0001

Data Index: 388252 (C4 (Web Text))

Max Activating Token Index: 640

Text #9

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1461. Min Act: -0.0000

Data Index: 1182067 (C4 (Web Text))

Max Activating Token Index: 169

Text #10

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1044. Min Act: -0.0001

Data Index: 274716 (C4 (Web Text))

Max Activating Token Index: 319

Text #11

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1306. Min Act: -0.0001

Data Index: 476364 (C4 (Web Text))

Max Activating Token Index: 986

Text #12

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1593. Min Act: -0.0001

Data Index: 605411 (C4 (Web Text))

Max Activating Token Index: 701

Text #13

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1658. Min Act: -0.0001

Data Index: 149264 (C4 (Web Text))

Max Activating Token Index: 39

Text #14

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1430. Min Act: -0.0001

Data Index: 289704 (C4 (Web Text))

Max Activating Token Index: 122

Text #15

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1260. Min Act: -0.0001

Data Index: 28619 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1346. Min Act: -0.0001

Data Index: 1097000 (C4 (Web Text))

Max Activating Token Index: 30

Text #17

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1296. Min Act: -0.0000

Data Index: 1500508 (Python Code)

Max Activating Token Index: 176

Text #18

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1029. Min Act: -0.0001

Data Index: 540251 (C4 (Web Text))

Max Activating Token Index: 190

Text #19

Max Range: 0.3847. Min Range: -0.3847

Max Act: 0.1263. Min Act: -0.0001

Data Index: 582489 (C4 (Web Text))

Max Activating Token Index: 895