Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3950 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.7304. Min Act: -0.0001

Data Index: 177705 (C4 (Web Text))

Max Activating Token Index: 14

Text #1

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6719. Min Act: -0.0001

Data Index: 1090064 (C4 (Web Text))

Max Activating Token Index: 252

Text #2

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6758. Min Act: -0.0001

Data Index: 1339385 (C4 (Web Text))

Max Activating Token Index: 149

Text #3

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6694. Min Act: -0.0001

Data Index: 1410054 (Python Code)

Max Activating Token Index: 33

Text #4

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6477. Min Act: -0.0001

Data Index: 875891 (C4 (Web Text))

Max Activating Token Index: 348

Text #5

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6387. Min Act: -0.0001

Data Index: 1236891 (C4 (Web Text))

Max Activating Token Index: 371

Text #6

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6590. Min Act: -0.0001

Data Index: 1030197 (C4 (Web Text))

Max Activating Token Index: 953

Text #7

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6326. Min Act: -0.0001

Data Index: 37675 (C4 (Web Text))

Max Activating Token Index: 658

Text #8

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6015. Min Act: -0.0001

Data Index: 394312 (C4 (Web Text))

Max Activating Token Index: 373

Text #9

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6262. Min Act: -0.0001

Data Index: 720579 (C4 (Web Text))

Max Activating Token Index: 42

Text #10

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6191. Min Act: -0.0001

Data Index: 827938 (C4 (Web Text))

Max Activating Token Index: 54

Text #11

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6410. Min Act: -0.0001

Data Index: 15796 (C4 (Web Text))

Max Activating Token Index: 95

Text #12

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6363. Min Act: -0.0000

Data Index: 338882 (C4 (Web Text))

Max Activating Token Index: 61

Text #13

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6129. Min Act: -0.0001

Data Index: 783743 (C4 (Web Text))

Max Activating Token Index: 591

Text #14

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6290. Min Act: -0.0001

Data Index: 955988 (C4 (Web Text))

Max Activating Token Index: 19

Text #15

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6041. Min Act: -0.0001

Data Index: 1181644 (C4 (Web Text))

Max Activating Token Index: 485

Text #16

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.5999. Min Act: -0.0001

Data Index: 479988 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.6230. Min Act: -0.0001

Data Index: 497858 (C4 (Web Text))

Max Activating Token Index: 91

Text #18

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.5640. Min Act: -0.0001

Data Index: 164889 (C4 (Web Text))

Max Activating Token Index: 421

Text #19

Max Range: 0.7304. Min Range: -0.7304

Max Act: 0.5742. Min Act: -0.0001

Data Index: 619983 (C4 (Web Text))

Max Activating Token Index: 791