Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1815 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.7314. Min Act: -0.0001

Data Index: 1151479 (C4 (Web Text))

Max Activating Token Index: 3

Text #1

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.6655. Min Act: -0.0001

Data Index: 1418174 (Python Code)

Max Activating Token Index: 220

Text #2

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.5639. Min Act: -0.0001

Data Index: 244585 (C4 (Web Text))

Max Activating Token Index: 223

Text #3

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.5165. Min Act: -0.0001

Data Index: 1117980 (C4 (Web Text))

Max Activating Token Index: 951

Text #4

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.5517. Min Act: -0.0001

Data Index: 458599 (C4 (Web Text))

Max Activating Token Index: 505

Text #5

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.4721. Min Act: -0.0001

Data Index: 1513697 (Python Code)

Max Activating Token Index: 431

Text #6

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.4615. Min Act: -0.0001

Data Index: 1381345 (Python Code)

Max Activating Token Index: 15

Text #7

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.4463. Min Act: -0.0001

Data Index: 667803 (C4 (Web Text))

Max Activating Token Index: 99

Text #8

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.4040. Min Act: -0.0001

Data Index: 1399463 (Python Code)

Max Activating Token Index: 633

Text #9

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.4318. Min Act: -0.0001

Data Index: 480171 (C4 (Web Text))

Max Activating Token Index: 45

Text #10

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.3497. Min Act: -0.0001

Data Index: 1062169 (C4 (Web Text))

Max Activating Token Index: 7

Text #11

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.3640. Min Act: -0.0001

Data Index: 1184069 (C4 (Web Text))

Max Activating Token Index: 10

Text #12

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.3510. Min Act: -0.0001

Data Index: 1135857 (C4 (Web Text))

Max Activating Token Index: 139

Text #13

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.3389. Min Act: -0.0001

Data Index: 1086520 (C4 (Web Text))

Max Activating Token Index: 883

Text #14

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.2965. Min Act: -0.0001

Data Index: 480605 (C4 (Web Text))

Max Activating Token Index: 56

Text #15

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.1758. Min Act: -0.0001

Data Index: 1579177 (Python Code)

Max Activating Token Index: 547

Text #16

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.3472. Min Act: -0.0001

Data Index: 1013446 (C4 (Web Text))

Max Activating Token Index: 169

Text #17

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.2667. Min Act: -0.0001

Data Index: 139701 (C4 (Web Text))

Max Activating Token Index: 161

Text #18

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.2959. Min Act: -0.0001

Data Index: 391118 (C4 (Web Text))

Max Activating Token Index: 313

Text #19

Max Range: 1.7314. Min Range: -1.7314

Max Act: 1.2935. Min Act: -0.0001

Data Index: 951815 (C4 (Web Text))

Max Activating Token Index: 5