Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 155 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.4137. Min Act: -0.0003

Data Index: 534382 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.3398. Min Act: -0.0003

Data Index: 1577090 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.3184. Min Act: -0.0003

Data Index: 1590192 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2694. Min Act: -0.0003

Data Index: 193632 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2981. Min Act: -0.0003

Data Index: 1390272 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2609. Min Act: -0.0003

Data Index: 638771 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2625. Min Act: -0.0003

Data Index: 91415 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2547. Min Act: -0.0003

Data Index: 277284 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2979. Min Act: -0.0003

Data Index: 1510886 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2311. Min Act: -0.0003

Data Index: 604458 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2129. Min Act: -0.0003

Data Index: 547799 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.2038. Min Act: -0.0003

Data Index: 832871 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1968. Min Act: -0.0003

Data Index: 1370514 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1945. Min Act: -0.0003

Data Index: 761245 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1681. Min Act: -0.0003

Data Index: 486031 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1672. Min Act: -0.0003

Data Index: 726878 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1553. Min Act: -0.0003

Data Index: 1490646 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1887. Min Act: -0.0003

Data Index: 256956 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1317. Min Act: -0.0003

Data Index: 1084034 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 1.4137. Min Range: -1.4137

Max Act: 1.1876. Min Act: -0.0003

Data Index: 700001 (C4 (Web Text))

Max Activating Token Index: 1023