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

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

Neuron 548 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.7208. Min Act: -0.0001

Data Index: 1609012 (Python Code)

Max Activating Token Index: 798

Text #1

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6075. Min Act: -0.0001

Data Index: 1184347 (C4 (Web Text))

Max Activating Token Index: 948

Text #2

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6082. Min Act: -0.0001

Data Index: 1540362 (Python Code)

Max Activating Token Index: 1022

Text #3

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6489. Min Act: -0.0001

Data Index: 1421232 (Python Code)

Max Activating Token Index: 423

Text #4

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6008. Min Act: -0.0001

Data Index: 1372496 (Python Code)

Max Activating Token Index: 599

Text #5

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6029. Min Act: -0.0001

Data Index: 1417214 (Python Code)

Max Activating Token Index: 965

Text #6

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5728. Min Act: -0.0001

Data Index: 1544090 (Python Code)

Max Activating Token Index: 807

Text #7

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5868. Min Act: -0.0001

Data Index: 1492848 (Python Code)

Max Activating Token Index: 282

Text #8

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5746. Min Act: -0.0000

Data Index: 289881 (C4 (Web Text))

Max Activating Token Index: 975

Text #9

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.6001. Min Act: -0.0001

Data Index: 1396832 (Python Code)

Max Activating Token Index: 540

Text #10

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5591. Min Act: -0.0001

Data Index: 1545956 (Python Code)

Max Activating Token Index: 181

Text #11

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5542. Min Act: -0.0001

Data Index: 352698 (C4 (Web Text))

Max Activating Token Index: 891

Text #12

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5272. Min Act: -0.0001

Data Index: 1647046 (Python Code)

Max Activating Token Index: 558

Text #13

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5314. Min Act: -0.0001

Data Index: 616811 (C4 (Web Text))

Max Activating Token Index: 51

Text #14

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5424. Min Act: -0.0000

Data Index: 647916 (C4 (Web Text))

Max Activating Token Index: 993

Text #15

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5216. Min Act: -0.0001

Data Index: 1603781 (Python Code)

Max Activating Token Index: 428

Text #16

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5188. Min Act: -0.0001

Data Index: 1431029 (Python Code)

Max Activating Token Index: 935

Text #17

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5515. Min Act: -0.0000

Data Index: 756960 (C4 (Web Text))

Max Activating Token Index: 822

Text #18

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5457. Min Act: -0.0000

Data Index: 552972 (C4 (Web Text))

Max Activating Token Index: 865

Text #19

Max Range: 0.7208. Min Range: -0.7208

Max Act: 0.5058. Min Act: -0.0001

Data Index: 1453824 (Python Code)

Max Activating Token Index: 759