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

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

Neuron 1570 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.9344. Min Act: -0.0001

Data Index: 1437951 (Python Code)

Max Activating Token Index: 767

Text #1

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.9985. Min Act: -0.0001

Data Index: 1399787 (Python Code)

Max Activating Token Index: 694

Text #2

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.9418. Min Act: -0.0001

Data Index: 1386576 (Python Code)

Max Activating Token Index: 285

Text #3

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.9558. Min Act: -0.0001

Data Index: 1505076 (Python Code)

Max Activating Token Index: 891

Text #4

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.9001. Min Act: -0.0001

Data Index: 598092 (C4 (Web Text))

Max Activating Token Index: 726

Text #5

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8555. Min Act: -0.0001

Data Index: 1579249 (Python Code)

Max Activating Token Index: 1016

Text #6

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8856. Min Act: -0.0001

Data Index: 1598852 (Python Code)

Max Activating Token Index: 456

Text #7

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8552. Min Act: -0.0001

Data Index: 1501737 (Python Code)

Max Activating Token Index: 110

Text #8

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8574. Min Act: -0.0001

Data Index: 1542567 (Python Code)

Max Activating Token Index: 222

Text #9

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8590. Min Act: -0.0001

Data Index: 1455079 (Python Code)

Max Activating Token Index: 568

Text #10

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8530. Min Act: -0.0001

Data Index: 1342954 (C4 (Web Text))

Max Activating Token Index: 852

Text #11

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8312. Min Act: -0.0001

Data Index: 1362161 (Python Code)

Max Activating Token Index: 705

Text #12

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8501. Min Act: -0.0001

Data Index: 1562880 (Python Code)

Max Activating Token Index: 505

Text #13

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.7678. Min Act: -0.0001

Data Index: 1546559 (Python Code)

Max Activating Token Index: 224

Text #14

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8160. Min Act: -0.0001

Data Index: 1531541 (Python Code)

Max Activating Token Index: 773

Text #15

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8400. Min Act: -0.0001

Data Index: 1513143 (Python Code)

Max Activating Token Index: 314

Text #16

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8412. Min Act: -0.0001

Data Index: 1422918 (Python Code)

Max Activating Token Index: 507

Text #17

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8609. Min Act: -0.0001

Data Index: 1376989 (Python Code)

Max Activating Token Index: 587

Text #18

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.8285. Min Act: -0.0001

Data Index: 1463485 (Python Code)

Max Activating Token Index: 842

Text #19

Max Range: 0.9985. Min Range: -0.9985

Max Act: 0.7949. Min Act: -0.0001

Data Index: 668411 (C4 (Web Text))

Max Activating Token Index: 248