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

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

Neuron 3584 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: 1.3633. Min Range: -1.3633

Max Act: 1.3552. Min Act: -0.0001

Data Index: 10945 (C4 (Web Text))

Max Activating Token Index: 272

Text #1

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.3633. Min Act: -0.0001

Data Index: 272283 (C4 (Web Text))

Max Activating Token Index: 980

Text #2

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1872. Min Act: -0.0001

Data Index: 212850 (C4 (Web Text))

Max Activating Token Index: 20

Text #3

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1081. Min Act: -0.0001

Data Index: 3471 (C4 (Web Text))

Max Activating Token Index: 620

Text #4

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1324. Min Act: -0.0001

Data Index: 1214944 (C4 (Web Text))

Max Activating Token Index: 34

Text #5

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1305. Min Act: -0.0001

Data Index: 916616 (C4 (Web Text))

Max Activating Token Index: 558

Text #6

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1135. Min Act: -0.0001

Data Index: 1033385 (C4 (Web Text))

Max Activating Token Index: 714

Text #7

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1041. Min Act: -0.0001

Data Index: 693961 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1942. Min Act: -0.0001

Data Index: 1068786 (C4 (Web Text))

Max Activating Token Index: 1007

Text #9

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1442. Min Act: -0.0001

Data Index: 967784 (C4 (Web Text))

Max Activating Token Index: 550

Text #10

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0729. Min Act: -0.0001

Data Index: 1269555 (C4 (Web Text))

Max Activating Token Index: 727

Text #11

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0611. Min Act: -0.0001

Data Index: 876185 (C4 (Web Text))

Max Activating Token Index: 578

Text #12

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1095. Min Act: -0.0001

Data Index: 824785 (C4 (Web Text))

Max Activating Token Index: 532

Text #13

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0696. Min Act: -0.0001

Data Index: 206495 (C4 (Web Text))

Max Activating Token Index: 142

Text #14

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0541. Min Act: -0.0001

Data Index: 898056 (C4 (Web Text))

Max Activating Token Index: 1009

Text #15

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0538. Min Act: -0.0001

Data Index: 1288576 (C4 (Web Text))

Max Activating Token Index: 12

Text #16

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.0548. Min Act: -0.0001

Data Index: 369084 (C4 (Web Text))

Max Activating Token Index: 126

Text #17

Max Range: 1.3633. Min Range: -1.3633

Max Act: 1.1377. Min Act: -0.0001

Data Index: 881255 (C4 (Web Text))

Max Activating Token Index: 749

Text #18

Max Range: 1.3633. Min Range: -1.3633

Max Act: 0.9882. Min Act: -0.0001

Data Index: 604563 (C4 (Web Text))

Max Activating Token Index: 817

Text #19

Max Range: 1.3633. Min Range: -1.3633

Max Act: 0.9819. Min Act: -0.0001

Data Index: 611234 (C4 (Web Text))

Max Activating Token Index: 317