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

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

Neuron 968 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.4316. Min Act: -0.0000

Data Index: 1519836 (Python Code)

Max Activating Token Index: 778

Text #1

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.3640. Min Act: -0.0000

Data Index: 1187590 (C4 (Web Text))

Max Activating Token Index: 688

Text #2

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.4803. Min Act: -0.0000

Data Index: 156477 (C4 (Web Text))

Max Activating Token Index: 247

Text #3

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.3689. Min Act: -0.0000

Data Index: 346656 (C4 (Web Text))

Max Activating Token Index: 194

Text #4

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1946. Min Act: -0.0000

Data Index: 1122273 (C4 (Web Text))

Max Activating Token Index: 783

Text #5

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.3160. Min Act: -0.0000

Data Index: 1649137 (Python Code)

Max Activating Token Index: 186

Text #6

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1642. Min Act: -0.0000

Data Index: 838139 (C4 (Web Text))

Max Activating Token Index: 135

Text #7

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.3427. Min Act: -0.0000

Data Index: 975567 (C4 (Web Text))

Max Activating Token Index: 664

Text #8

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1867. Min Act: -0.0000

Data Index: 1135690 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1714. Min Act: -0.0000

Data Index: 1445873 (Python Code)

Max Activating Token Index: 214

Text #10

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.2870. Min Act: -0.0000

Data Index: 1550479 (Python Code)

Max Activating Token Index: 716

Text #11

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1433. Min Act: -0.0000

Data Index: 1087166 (C4 (Web Text))

Max Activating Token Index: 374

Text #12

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1512. Min Act: -0.0000

Data Index: 1440142 (Python Code)

Max Activating Token Index: 275

Text #13

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1319. Min Act: -0.0000

Data Index: 680946 (C4 (Web Text))

Max Activating Token Index: 118

Text #14

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1319. Min Act: -0.0000

Data Index: 848769 (C4 (Web Text))

Max Activating Token Index: 967

Text #15

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.0651. Min Act: -0.0000

Data Index: 1282162 (C4 (Web Text))

Max Activating Token Index: 265

Text #16

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.0340. Min Act: -0.0000

Data Index: 1188724 (C4 (Web Text))

Max Activating Token Index: 633

Text #17

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1570. Min Act: -0.0000

Data Index: 294881 (C4 (Web Text))

Max Activating Token Index: 273

Text #18

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1309. Min Act: -0.0000

Data Index: 1491281 (Python Code)

Max Activating Token Index: 950

Text #19

Max Range: 2.4803. Min Range: -2.4803

Max Act: 2.1264. Min Act: -0.0000

Data Index: 1193710 (C4 (Web Text))

Max Activating Token Index: 575